ISSN 1055- 1425
November 2007
This work was performed as part of the California PATH Program of the
University of California, in cooperation with the State of California Business,
Transportation, and Housing Agency, Department of Transportation, and the
United States Department of Transportation, Federal Highway Administration.
The contents of this report reflect the views of the authors who are responsible
for the facts and the accuracy of the data presented herein. The contents do not
necessarily reflect the official views or policies of the State of California. This
report does not constitute a standard, specification, or regulation.
Final Report for Task Order 6216
CALIFORNIA PATH PROGRAM
INSTITUTE OF TRANSPORTATION STUDIES
UNIVERSITY OF CALIFORNIA, BERKELEY
Expedited Crash Investigation - With Use
of Technologies for Documentation and
Processing
UCB- ITS- PRR- 2007- 18
California PATH Research Report
Ching- Yao Chan
Thang Lian
Jeff Ko
CALIFORNIA PARTNERS FOR ADVANCED TRANSIT AND HIGHWAYS
Final Report for
Task Order 6216
EXPEDITED CRASH INVESTIGATION
- WITH USE OF TECHNOLOGIES FOR DOCUMENTATION AND PROCESSING
Ching- Yao Chan, Thang Lian, Jeff Ko
California PATH
Institute of Transportation Studies
University of California at Berkeley
April 30, 2007
i
ii
ABSTRACT
The mobility and efficiency of California highways are impeded by the recurrent and non-recurrent
congestion on a daily basis. Roadway incidents, especially collisions, often result in
traffic congestion and travel delays. This project is initiated to explore the use of technologies
that will potentially bring direct and immediate benefits to the law enforcement officers and
other personnel who are involved in the handling of collision sites and subsequent investigations.
The work carried out in this project includes three major components: ( 1) The experimentation of
integrated vehicular technology systems to provide assistance for law enforcement officers, ( 2)
The exploration of utilizing GPS devices for vehicle and incident locations, and ( 3) The
development of photogrammetry tool to extract supplementary information from collision
scenes.
Valuable insights were gained through collaborations with the University of New Hampshire
Project 54 and the City of Carlsbad, California, Public Safety Technology Programs. The key
factors to successful deployment of technological systems are the easy- to- learn operation
features and friendliness of user- machine interface. Integrated multiple- function capabilities and
open architecture are important considerations in adopting technological systems to ensure
favorable cost- benefit returns of investments from the perspectives of local jurisdictions or
agencies.
KEY WORDS
GPS, Driver- Assistance Systems, Technology for Law Enforcement, Project 54, Public Safety,
Collision Database, Geo- Coding, Photogrammetry
iii
iv
TABLE OF CONTENTS
ABSTRACT....................................................................................................................... ......... III
KEY WORDS.......................................................................................................................... ... III
LIST OF FIGURES.................................................................................................................. VII
IX
112
456889
10
11
12
12
13
13
14
15
16
16
16
18
19
20
20
21
22
23
25
25
25
26
27
28
33
EXECUTIVE SUMMARY ........................................................................................................
1. BACKGROUND...................................................................................................................
1.1 Research Objectives..................................................................................................
1.2 Research Approaches ...............................................................................................
2. GPS TECHNOLOGIES....................................................................................................... 3
2.1 GPS Basics ................................................................................................................. 3
2.2 US Nationwide Higher Accuracy GPS.................................................................... 3
2.3 Mapping and GIS......................................................................................................
2.4 GPS- Enabled Functionalities for Crash Investigation and Documentation .......
2.5 Synergistic Research Activities in California.........................................................
3. INTEGRATED ON- BOARD SYSTEMS FOR LAW- ENFORCEMENT USERS ........
3.1 About Project 54 and CATlab at University of New Hampshire .........................
3.2 Integrated In- Vehicle Functions and Dispatcher Center – City of Carlsbad .....
3.2.1 Carlsbad Experience in Public Safety technology Project............................
3.2.2 Latest Status of Carlsbad Technology Implementation and Feedback.........
3.3 Experimental Evaluation at University of California at Berkeley .....................
3.3.1 GPS data recording .......................................................................................
3.3.2 GPS Navigation ............................................................................................
3.3.3 GPS Data Collection Format ........................................................................
3.3.4 GPS Data Display and Mapping...................................................................
3.3.5 Summary of UCB- PD Project 54 and GPS Evaluation ................................
4. PHOTOGRAMMETRY.....................................................................................................
4.1 Photogrammetry Basics..........................................................................................
4.2 Preliminary Version of Photogrammetry Application........................................
4.3 Revised and Enhanced Version of Photogrammetry Application......................
4.3.1 Revised Screen Display and User Interface..................................................
4.3.2 Aerial Photo Option ......................................................................................
4.3.3 Gird Adjustment Options..............................................................................
4.3.4 Reference Square ..........................................................................................
4.4 Accuracy of Photogrammetry Application...........................................................
4.5 Future Extensions ...................................................................................................
5. SUMMARY AND CONCLUSIONS .................................................................................
5.1 Scope of Technology Survey and Experimentation .............................................
5.2 Summary of Findings .............................................................................................
5.3 Recommendations ...................................................................................................
REFERENCES..................................................................................................................... ......
APPENDIX A: EXPERIMENTAL VEHICLE INSTRUMENTATION ..................................
APPENDIX B: CAMERA MODEL FOR SPATIAL PROJECTION IN
PHOTOGRAMMETRY...............................................................................................................
v
vi
LIST OF FIGURES
Figure 1 In- Vehicle Components in Police Car Passenger Compartment .............................. 9
10
12
13
14
15
17
18
19
21
22
23
23
Figure 2 Equipment and Device Interface in Police Car Truck Space..................................
Figure 3 GPS Logger Graphic User Interface .........................................................................
Figure 4 Mapping Software Interface.......................................................................................
Figure 5 GPS Data Coordinate Display...................................................................................
Figure 6 GPS Data Display Overlapping with Local Street Map ..........................................
Figure 7 Basic Screenshot of Photogrammetry Application...................................................
Figure 8 Screenshot with Multiple Defined Points on the Photo View and Corresponding
Orthographic View .....................................................................................................................
Figure 9 Basic Screenshot of Revised Photogrammetry Application ....................................
Figure 10 Example of A Reference Square...............................................................................
Figure 11 Image Panel after Grid Generation .........................................................................
Figure 12 Measurement of a Tape Measure.............................................................................
Figure 13 Application Calculating Tape Length .....................................................................
vii
viii
EXECUTIVE SUMMARY
The mobility and efficiency of California highways are impeded by the recurrent and non-recurrent
congestion on a daily basis. Roadway incidents, especially collisions, often result in
traffic congestion and travel delays. Such congestion is mainly caused by stopped vehicles and
lane closure, but it is intensified by slowing vehicles with drivers observing the accident scene.
The effects of these phenomena on highway traffic are significant in terms of their direct hazards
and the associated losses in travel delays, energy usage, and environmental impact.
This project was initiated to explore the use of technologies that will potentially bring direct and
immediate benefits to the law enforcement officers and other personnel who are involved in the
handling of collision sites and subsequent investigations. The work carried out in this project
includes three major components: ( 1) The experimentation of integrated vehicular technology
systems to provide assistance for law enforcement officers, ( 2) The exploration of utilizing GPS
devices for vehicle and incident locations, and ( 3) The development of photogrammetry tools to
extract supplementary information from collision scenes.
During the course of the project, the research team carried out a broad survey and assessed the
availability and suitability of individual devices and packaged systems for law enforcement
applications. After initial evaluation, a collaborative relationship was established with the
University of New Hampshire CATlab, where Project54 ™ was developed. This cooperative
effort allowed us immediate access to the mature and already deployed technology set that
appears to be an ideal candidate to be considered for law enforcement and state agency vehicle
fleets. With its wide deployment in a large number of police vehicles, Project 54 has evolved
into an attractive driver- assistance system. It main features are:
( 1) The system offers multiple user interfaces by voice, touch- screen, conventional switches or
knobs to allow the users to activate on- board devices under various field operating conditions,
( 2) The use of different user interfaces is not mutually exclusive and thus each individual user
can opt for appropriate control methods at different times or under different conditions,
( 3) The system has a short learning curve due to its user- friendliness,
( 4) The system has a non- proprietary open architecture, which allows the flexible selection and
replacement of modular components and sub- systems.
The research team also interacted with the city of Carlsbad Police Department, where a Public
Safety Technology Problem was implemented with great success. Considerable resources were
dedicated to implement infrastructure and vehicle renovations to provide officers with mobile
office capability, including:
( 1) Registration/ Stolen car inquiries,
( 2) Driver license checks,
( 3) Wanted persons checks,
( 4) Email, Reports, and Pictures,
( 5) Crime analysis,
( 6) Access all city and county justice systems.
Subsequently, the research team made arrangements with the University of California, Berkeley,
Police Department and initiated the instrumentation of an experimental vehicle on one UCB- PD
ix
x
cruiser. The objectives of the experiments were to assess the following two aspects in
accordance with the research plans:
( 1) Exploring the technical and institutional issues in local environment when an officer-assistance
system such as Project 54 is installed,
( 2) Evaluating the use of vehicle- mounted GPS device for the purpose of vehicle location
identification and incident reporting.
The results of GPS data experiments and Project 54 at UCB- PD can be summarized as follows:
( 1) Low- cost GPS devices are readily available with decent resolutions and performance
specifications for the purpose of incident reporting,
( 2) GPS Data recording in the Berkeley neighborhood, a suburban region with scattered tall
buildings and occasional narrow streets, was consistent and reliable for its intended usage,
( 3) Mapping and navigation was a function preferred by officers during the testing,
( 4) Project 54 is an easy- to- use system with minimum learning curves. However, the scale of
experimentation with one single vehicle was too small to fairly assess its benefits,
( 5) The radio units in the UCB- PD cruiser were not compatible with Project 54, as a result, the
overall performance was significantly reduced,
( 6) More training and user interactive sessions are strongly desirable.
One primary factor in the delays of incident handling is the necessary and proper documentation
of incident information for legal and technical reasons. The prevailing method of scene
measurement is often conducted by physically walking the scenes with wheel measures or
similar tools. The common method and its conventional approach present several issues: time
consumption, road hazard, limitations at site, and needs for photographic evidence. In this
project, we investigated and enhanced a computer software application using photogrammetric
techniques to facilitate the measurement of the aforementioned incident scenes. This software
will allow the user to simply take the photographs taken at the crash sites and extract
supplementary measurements in the convenience of the office, solving the shortcomings of the
conventional method.
1. BACKGROUND
The mobility and efficiency of California highways are impeded by the recurrent and non-recurrent
congestion on a daily basis. Roadway incidents, especially collisions, often result in
traffic congestion and travel delays, in addition to the direct damage to the vehicles and the
injuries to the people involved. The subsequent congestion is mainly caused by stopped vehicles
or lane closure, but it is intensified by slowing vehicles with drivers observing the accident scene.
The effects of these phenomena on highway traffic are significant in terms of their direct hazards
and the associated losses in travel delays, energy usage, and environmental impact.
This project was initiated to explore the use of technologies that will potentially bring direct and
immediate benefits to the patrol officers and operators that are involved in the handling of
collision sites and subsequent investigations. The work is based on and extended from previous
research that was developed under the sponsorship of California Office of Traffic Safety ( OTS)
and the collaboration of California Highway Patrol ( CHP) with an emphasis on the deployment
of technical tools. [ 1, 2]
One primary objective of the project is to define user specifications and useful functionalities for
the use of Global Positioning System ( GPS) that can facilitate expedient documentation of
collision sites. The availability of GPS data can then be further incorporated into the data entry
process and statistical analysis in collision databases such as SWITRS ( Statewide Integrated
Traffic Records System) [ 3] and TASAS ( Traffic Surveillance Accident Surveillance and
Analysis System) [ 4]. The availability of GPS data in collision reports and database will allow
effective record keeping and to enable safety performance evaluation of the state- wide highway
network.
1.1 Research Objectives
In the last ten years, on average more than 3,500 fatality, 200,000 injury, and 300,000 property-damage
collisions occurred on California roadways every year [ 3]. The direct costs of these
collisions are enormous, yet their impacts on highway efficiency are even more daunting
considering all the resources needed to handle the consequences of roadway crashes. First of all,
law enforcement officers, assisting transport vehicle and paramedics are required to attend to the
people and vehicles involved in the collisions. Secondly, traffic near the crash scenes queues up
quickly and often leads to congestion that may take a considerable period of time to dissipate. It
is not uncommon for a single incident during the rush hours to affect traffic conditions for the
whole morning or afternoon. The associated losses in personnel time, congestion delays, energy
usage, environmental impacts, and secondary collisions can certainly be considered as one major
culprit that leads to the deterioration of highway mobility, efficiency, and safety on a daily basis.
Marginal improvements in the response and handling of crashes can result in significant benefits
towards mitigating this complex problem, in spite of the enormity of the problems.
The primary objectives of this project are two- fold:
( 1) Development of GPS and vehicle on- board instrumentation
1
The first aspect of efficiency improvements will come from the ease and expediency of handling
various tasks in the field by law enforcement officers. Specifically, this project continues
previous developments [ 1, 2] and focuses on the following options:
􀁺 Exploring integration of on- board or mobile technologies that incorporates the use of GPS,
bar code and magnetic stripe reader, voice recorder, and other commercial- off- the- shelf
products that can offer assistance for the process of information collection and incident
handling in the field.
􀁺 Identifying user preferences and deployment options of GPS units that can enable the
expedient documentation of collision locations.
( 2) Software for post- processing of photographic evidence
The other improvement in crash investigation relies on the capability of examining photographs
to gather additional evidence, which are either unavailable or incomplete at the time of on- site
documentation. Thus, one major task in the project is to continue the development of a
photogrammetry tool and to enhance functionalities on a preliminary version of such software
from earlier work. [ 2]
1.2 Research Approaches
Based on the work of an earlier project sponsored by OTS [ 1, 2], the research team has
established a baseline understanding of various technology devices that may be applicable for
crash investigation. More importantly, through the interaction with CHP officers, we also learned
about the user needs and institutional issues that are critical for field operations by law-enforcement
personnel. Thus, we defined and followed a research plan with the following
guidelines:
( 1) Engage the users in an iterative process to ensure that the development and testing of
suggested systems will be user friendly and acceptable to the intended target audience,
( 2) Although stand- alone functions or devices may serve particular purposes, an integrated
system that incorporates a wide choice of features can best meet the diverse needs of
individual users,
( 3) With the fast advancements in various technological fields, there are a great variety of
commercially off- the- shelf products that can be modularly adopted and flexibility integrated
without unnecessarily high development costs. . Thus, the hardware and software
architecture should remain open and allow flexibility in component or sub- system redesign
and replacements,
( 4) Collaboration with other institutions should be a priority to leverage off previous research
and existing resources,
( 5) When evaluating the effectiveness of deployable solutions, the involvement of the local
government agencies and relevant institutional issues should be taken into account.
2
2. GPS TECHNOLOGIES
One primary objective of the project is to seek solutions for the use of Global Positioning System
( GPS) devices that can facilitate expedient documentation of collision sites. The availability of
GPS data can then be further incorporated into the data entry process and statistical analysis in
collision databases such as SWITRS ( Statewide Integrated Traffic Records System) [ 3] and
TASAS ( Traffic Surveillance Accident Surveillance and Analysis System) [ 4]. The availability
of GPS data in collision reports and database will allow effective record keeping and enable
safety performance evaluation of the state- wide highway network.
With decreasing prices and increasing market penetration of Global Positioning System ( GPS) in
vehicle navigation markets and hand- held devices, GPS has gradually become a familiar
household item. Furthermore, the availability of higher precision measurements by GPS will
potentially spread broader implementation of advanced applications for private users as well as
public agencies. This section provides a review of GPS- related technologies applicable to the
subject areas of this project.
2.1 GPS Basics
The Global Positioning System ( GPS) is a worldwide radio- navigation system formed from a
constellation of 24 satellites and their ground stations. [ 5] GPS uses these " man- made stars" as
reference points to calculate positions accurate to a matter of meters. In fact, with advanced
forms of GPS you can make measurements to better than a centimeter.
The quest for greater and greater accuracy has spawned an assortment of variations on basic GPS
technology. One technique, called " Differential GPS," involves the use of two ground- based
receivers. One monitors variations in the GPS signal and communicates those variations to the
other receiver. The second receiver can then correct its calculations for better accuracy.
Another technique called " Carrier- phase GPS" takes advantage of the GPS signal's carrier
signal to improve accuracy. The carrier frequency is much higher ( which also implies a much
shorter wavelength) than the GPS signal which means it can be used for more precise timing
measurements.
The aviation industry is developing a type of GPS called " Augmented GPS" which involves the
use of a geostationary satellite as a relay station for the transmission of differential corrections
and GPS satellite status information. These corrections are necessary if GPS is to be used for
instrument landings. The geostationary satellite would provide corrections across an entire
continent.
2.2 US Nationwide Higher Accuracy GPS
Currently, the GPS service offers a 4- to 20- meter ( m) navigational accuracy. For many land
transportation uses, this accuracy is insufficient. The Nationwide Differential Global Positioning
System ( NDGPS) offers a 1- to 3- m radio- navigational service that meets the needs of many
more transportation users. [ 6]
3
The High Accuracy- Nationwide Differential Global Positioning System ( HA- NDGPS) program
[ 7] provides the capability to broadcast corrections to the Global Positioning System ( GPS) over
long ranges to achieve a better than 10 centimeter ( cm) ( 95 percent) accuracy throughout the
coverage area. HA- NDGPS is currently undergoing a research and development phase. The
signal is available for test purposes from Hagerstown, MD, and soon Hawk Run, PA.
Application of this technology will provide advanced safety features for transportation, including
lane departure warning, intersection collision warnings, and railroad track defect alerts. It also
could be used for economic enhancements such as precision container tracking and automated
highway lane striping.
Because greater precision is needed to support many of the safety enhancements envisioned for
the future, the U. S. Department of Transportation, in conjunction with the Interagency GPS
Executive Board, is supporting the development of HA- NDGPS to provide 10 cm horizontal and
20 cm vertical ( 95 percent) corrections to users. HA- NDGPS uses the infrastructure employed by
the NDGPS service to broadcast these corrections. The addition of a diplexer and transmitter
allow the existing infrastructure to broadcast the additional signal, keeping implementation costs
very low. Additionally, the signal will be monitored to ensure it is providing the accuracy needed
to meet safety- of- life applications.
In separate Caltrans- sponsored activities ( VII California – PATH Task Order 6217), California
DOT is evaluating the applicability of HA- NDGPS for a variety of safety and traffic
management functions through the establishment of a high- accuracy base station in California.
[ 8- 10] Once this facility becomes available, interested parties can implement applications that
require very high- accuracy positioning resolutions.
2.3 Mapping and GIS
A geographic information system ( GIS) is a system for capturing, storing, analyzing and
managing data and associated attributes which are spatially referenced to the earth. In the
strictest sense, it is a computer system capable of integrating, storing, editing, analyzing, sharing,
and displaying geographically- referenced information. In a more generic sense, GIS is a tool that
allows users to create interactive queries ( user created searches), analyze the spatial information,
edit data, maps, and present the results of all these operations. Geographic information science is
the science underlying the applications and systems, taught as a degree program by several
universities.
Geographic information system technology can be used for scientific investigations, resource
management, asset management, Environmental Impact Assessment, Urban planning,
cartography, criminology, history, sales, marketing, and route planning. For example, a GIS
might allow emergency planners to easily calculate emergency response times in the event of a
natural disaster, a GIS might be used to find wetlands that need protection from pollution, or a
GIS can be used by a company to find new potential customers similar to the ones they already
have and project sales due to expanding into that market.
4
For the purposes of this project, the application of GIS is mainly related to the identification of
collision locations. This can be utilized at two different levels:
( 1) Real- time tracking of incidents
For law enforcement and traffic management functions, the missions are often time critical in
responding and handling the events. For crash investigation, this means the dispatch of officers
and paramedics must be routed to the scene within the shortest time possible. The GPS
coordinates of patrol vehicles, when integrated with GIS, will enable a clear and rapid
identification of their locations relative to incident sites. Once the location of vehicles can be
seen by traffic management centers or dispatch centers, additional support from nearby patrol
areas or jurisdictions can also be sent expediently. Furthermore, the reporting and
documentation of site locations can be carried out easily with the activation of an on- board
function by the onsite personnel. The data can then be sent wirelessly and recorded locally when
desired.
( 2) Post- processing
If the site information ( such as GPS coordinates) can be automatically incorporated into the
reports of individual incidents and collisions, an information- rich database can be gradually and
increasingly established. The availability of such database will then enable powerful
identification, search, and analysis of collision database, which can also be linked to a
customized GIS for roadway network management.
2.4 GPS- Enabled Functionalities for Crash Investigation and Documentation
GPS is a powerful enabling technology. Specifically, its application in crash investigation and
documentation includes the following categories of functions:
( 1) Automatic Vehicle Location ( AVL) by providing a message from the vehicles in real time.
( 2) Emergency Response by locating the positions of vehicles and incidents.
( 3) Site coordinate recording for crash locations.
A very relevant study was conducted in Kentucky on the use of GPS devices to record GPS
coordinates for accident reports. [ 11] GPS equipment and training were provided to all police
agencies throughout the state. The study was to evaluate the accuracy of this technology in
locating traffic crashes which would be critical to having an effective safety program. The major
findings from the study were:
( 1) The GPS devices were capable of providing accurate location of a crash site.
( 2) Substantial differences were found between the locations of some crashes with GPS vs. mile
point ( CRMP) data.
( 3) GPS data was somewhat more accurate than CRMP information.
( 4) Errors were typically operator errors rather than the equipment problems.
( 5) Training and some hardware improvements would solve most of the problems encountered.
( 6) The police report should be modified to reflect proper GPS data format.
From the findings of this recent study, it can be seen that user training and instructions are most
critical in obtaining the correct information for the purposes of collision location recordings.
Even though most updated handheld GPS devices can probably achieve the same level of
accuracy as vehicle- mounted GPS ones, due to the vulnerability to entry errors they should only
be used selectively for situations where the use of vehicular- based systems is limited.
5
Furthermore, a GPS device integrated into a vehicle on- board computing system with automated
data entry options is likely to provide more consistent and reliable recordings.
2.5 Synergistic Research Activities in California
Under a project sponsored by the Office of Traffic Safety ( OTS), the Traffic Safety Center at the
University of California at Berkeley is undertaking an effort for the geo- coding of collision
database. [ 12] A summarized description of the project and the related issues is given below.
Each year local and governmental agencies collect and analyze California traffic collision data
( SWITRS, Statewide Integrated Traffic Records System) to monitor injury rates, identify high
collision locations, develop traffic safety programs, and evaluate the effectiveness of safety
measures. Many SWITRS data users need to link motor vehicle collision data with exact
geographical information to identify dangerous roads, intersections, and to study crash patterns
on specific road and intersection types. There are currently many barriers to accurate,
inexpensive, and efficient means of accesses geo- coded collision data. They are:
( 1) Expense: Commercially available platforms to geo- code SWITRS data are very expensive.
( 2) Ease- of- use: Commercially available geo- coding engines generally use a single data field to
match addresses, and occasionally use a secondary zone field ( e. g., zip code, city, county) to
prevent out- of- area matches. Current location information in SWITRS, however, is
represented by a collection of data fields including primary and secondary roads, qualified by
direction and offset fields. Therefore, special programming is needed to precisely geo- code
SWITRS data into commercially available software.
( 3) Inaccuracy: Accurate geo- coding requires the use of consistent street names, correctly spelled
street names, accurate “ offset” and “ direction” data fields estimated by the reporting officer,
and a current and extremely accurate area map. Due to these barriers, most geo- coding is
inadequate for use in analysis of intersection safety.
( 4) Inefficiency: Even with the best software, programming and base map, some crashes will
require manual geo- coding, a very labor intensive process.
( 5) Redundancy: Many individual jurisdictions, county jurisdictions and some state jurisdictions
are currently geo- coding crash data. The geo- coding being done may be duplicated by other
researchers unaware of the overlap.
Until first responders use the Global Positioning System ( GPS) to record the location of a crash,
geo- coding crash location is critical for researchers and local communities to map collision
occurrences. A centralized effort to provide accurate coordinates for geo- coded crashes would
resolve current impediments to traffic safety research and put the State of California at the
forefront of technological solutions for public health.
Currently, the California Highway Patrol is investigating possibilities for the automatic inclusion
of GPS collision locations in the SWITRS data. The data can be geo- coded in two fashions. The
first possibility is to equip all first- responders with GPS units and require them to report the GPS
location on each collision report form. The second approach is for the CHP to use available
location information to extrapolate the GPS coordinates of each collision. The former approach
requires a one- time but tremendous overhaul of the collision reporting process; the later
approach requires the CHP to commit to a yearly effort to geo- code all of the data using special
software programs. In the current state of technology, no software program can produce 100%
6
accurate estimates, and most often software is unable to produce any estimate for a significant
fraction of the collisions ( 15- 20%). The resulting “ mismatches” then require significant, often
manual, attention. The CHP, and other states’ highway agencies, are actively researching
solutions to this dilemma. Other local agencies have begun, on a piecemeal basis, to geo- code
data.
7
3. INTEGRATED ON- BOARD SYSTEMS FOR LAW- ENFORCEMENT USERS
In the beginning phase of the project, the research team carried out a broad- based survey and
assessed the availability and applicability of individual devices and sub- systems for law
enforcement applications. Subsequently, the PATH researchers established a collaborative
relationship with Professor Andrew Kun of the University of New Hampshire ( UNH) CATlab,
where Project54 ™ was developed. This cooperative effort allowed the project to have
immediate access to a promising and field- tested technology package that appeared to be an ideal
candidate suitable for law enforcement and state agency vehicle fleets. In addition, the project
team also solicited inputs and received advices from Captain Dale Stockton, now retired, of the
City of Carlsbad Police Department in California, where Project54 and other related technology
systems are implemented. Through interaction with Carlsbad and UNH, considerable insights
were gained in users’ feedback as well as the keys to success of an integrated public safety
program.
3.1 About Project 54 and CATlab at University of New Hampshire
The CATlab project is a collaborative research and development effort between the University of
New Hampshire and the New Hampshire Department of Safety and is supported by the U. S. [ 13]
Department of Justice, through the effort and continued support of Senator Judd Gregg. The
faculty and students of CATlab work on introducing advanced technologies into the operations
of the New Hampshire State Police and other law enforcement agencies.
Today's cruisers are equipped with digital radios, GPS units, computers, radars, lights, sirens, etc.
Dispatch centers have become computerized, and officers on the beat and in offices access a
variety of databases on a daily basis. However, these devices are most often not designed to
become a part of a system of multiple devices manufactured by different companies. In other
words they are not designed with integration in mind.
The integration of devices and systems is the primary interest of CATlab. Most of the work is
done on integrating electronic devices in police cruisers. The integrated Project54 system allows
officers to interact with equipment such as lights and siren, radar, etc. using speech input and
feedback. The Project54 system also integrates police cruisers into state- wide data networks.
As of April 19, 2007, 771 law enforcement vehicles are equipped with Project54 ™ system in NH
and 156 vehicles out- of- state are completed.
With its wide deployment in a large number of police vehicles, Project54 has evolved into a
sophisticated and an attractive driver- assistance system. It main features are:
( 1) The system offers multiple user interfaces by voice, touch- screen, conventional switches or
knobs to allow the users to activate on- board devices under various field operating conditions.
( 2) The use of different user interfaces is not mutually exclusive and thus each individual user
can opt for appropriate control methods at different times or under different conditions.
( 3) The system has a short learning curve due to its user- friendliness.
( 4) The system has a non- proprietary open architecture, which allows the flexible selection and
replacement of modular components and sub- systems.
8
3.2 Integrated In- Vehicle Functions and Dispatcher Center – City of Carlsbad
The Carlsbad, California Police Department undertook a major technology project that was
designed to provide true mobile office capability to officers in the field. [ 8] In short, the police
car’s computer allows the officer full access to the traditional law enforcement data bases and
dispatch information as well as full access to the city’s network resources, email and internet
functions. Using a wireless infrastructure based on CDMA EVDO technology, the project
permits broadband capability from the patrol car, thus supporting a greater variety of applications.
As the project design progressed, consideration was given to enhancing officer capability and
safety by using voice recognition.
After extensive inquiry, the department learned of the Project54 effort. After initial evaluation,
the Carlsbad technology venture was expanded to include Project54 equipment. Carlsbad PD
worked with UNH to modify the Project54 software source code to perform California DMV
inquiries. Carlsbad’s use of the Project54 software marks the first use of the technology outside
the state of New Hampshire.
Figures 1 and 2 below depict some components of the Carlsbad Technology implementation and
installation.
Figure 1 In- Vehicle Components in Police Car Passenger Compartment
9
Figure 2 Equipment and Device Interface in Police Car Truck Space
3.2.1 Carlsbad Experience in Public Safety technology Project
Captain Dale Stockton [ 14] highly praised the effectiveness of Project54 after his department
implemented such systems on 50 of their patrol cruisers. He pointed out that the keys to success
of Project54 consist of the following factors:
􀁺 Simple non- proprietary interfaces,
􀁺 Compatibility with the widest possible range of equipment,
􀁺 Single site license for $ 500 to cover an agency regardless of the number of vehicles,
􀁺 Flexible and easy adaptation of desired components,
􀁺 Specialized digital array microphone for voice recognition,
􀁺 Short learning curve for new users,
􀁺 Great cost- benefit returns for agencies.
In addition to the adoption of Project54, Carlsbad’s successful Public Safety Technology
program can be contributed to the additional renovation of the complete Information Technology
Architecture. They worked with a vendor to implement the following components and functions:
􀁺 Cellular Wireless Communication Infrastructure, ( This item was not bulleted in the article)
􀁺 Mapping- GIS,
􀁺 Automatic Vehicle Locator,
􀁺 Record Management,
10
􀁺 Automated field reporting.
The overall system results in a mobile office capability for officers, which allows the officers to
perform a variety of functions, including:
􀁺 Registration/ Stolen car inquiries,
􀁺 Driver license checks,
􀁺 Wanted persons,
􀁺 Email,
􀁺 Reports,
􀁺 Pictures,
􀁺 Crime analysis,
􀁺 Access all city and county justice systems.
Carslbad PD indicates that AVL has been an effective resource management tool and it improves
services to the citizens by reducing response time in critical incidents. Importantly, the system
offers safety protection to officers during the execution of field operations beside the
convenience in lessening the driver/ officer workloads.
3.2.2 Latest Status of Carlsbad Technology Implementation and Feedback
The research team made several visits to Carlsbad and communicated with the management as
well as officers who used the technology system in the field. Some highlights are summarized as
follows:
􀁺 Previously the GIS- linked mapping capability was only implemented in the dispatcher
center. It is now also available in the police cruiser so that officers can check the locations
of reported incidents as well as others officers in the neighborhood.
􀁺 License plate inquiry is fully functional. Carlsbad PD uses Verizon Wireless
Communication as the telecommunication supplier. The inquiry data is sent through a
Verizon wireless modem on board when officers made the request. The data is sent through
the city office server, then subsequently to the county and state data server. The return
message is sent in the reverse route and arrives in the police car on the monitor screen.
􀁺 The license check function can be done by using the conventional cruiser radio, the
computer keyboard, or the voice recognition interface of Project54.
􀁺 The driver license check function is more challenging in California due to the large number
of people who may have the same names as others and the diversity of names from different
ethnic groups.
􀁺 Under certain circumstances, it is best that officers keep their eyes on the suspects. A return
to their cruisers to perform other functions may put the officers in danger. Therefore, it will
be desirable if a wireless bar- code reader or an externally accessible magnetic reader can
transmit the data into the cruiser for inquiries.
􀁺 The researcher team was provided with a chance to ride along with officers. One officer
just had the license plate check system installed on that day. His response was “ awesome,”
and indicated that he would exclusively use the voice- activation for license check in the
future. This showed that there were few learning obstacles despite that one particular
officer might be more technology inclined than others.
11
􀁺 The officer was particularly excited about the opportunity to use the advanced computerized
system. He commented that the major advantage, besides the ease of use in the field, was
the safety benefit for the officers to keep their eyes and attention more on the road and on
the people that they need to watch.
􀁺 Meaningfully, based on communication with officers on duty who used the Public Safety
Technology system, the system has been received favorably with very positive overall user
experience.
3.3 Experimental Evaluation at University of California at Berkeley
The research team established collaborative relationships with UNH and Carlsbad after we
learned of the success and advanced developments of project 54 Systems. Initially, the research
team offered to provide a Project 54 system to CHP for their evaluation. However, due to the
internal planning issues with CHP, the plan was halted. Subsequently, the research team
contacted the University of California Berkeley Police Department and initiated the
instrumentation of an experimental vehicle on one UCB- PD cruiser. The objectives of the
experiments were to assess the following two aspects in accordance with the research plans:
( 1) Exploring the technical and institutional issues in local environment when an officer-assistance
system such as Project 54 is installed.
( 2) Evaluating the use of vehicle- mounted GPS device for the purpose of vehicle location
identification and incident reporting.
The detailed descriptions of various components installed in the experimental vehicle at UCB-PD
are given in Appendix A.
3.3.1 GPS data recording
Figure 3 GPS Logger Graphic User Interface
12
For the purpose of GPS evaluation, the research team developed a data logger component on the
experimental vehicle at Berkeley. GPS data recording is implemented in the Project 54 using
C++. The data is recorded every second each time the GPS GUI is updated. Data recording
starts at the beginning of the system every time the vehicle is started. Figure 3 shows the GPS
Logger GUI of the system.
3.3.2 GPS Navigation
The UCB- PD requested that mapping software be installed on the experimental vehicle, as the
officers indicated that navigation would be a preferred function on the vehicle. A mapping
software, CoPilot, was later installed for the navigation system ( Figure 4). The GPS data is
received through the antenna mounted on the back of the trunk. USB GPS interface is connected
through a 4- port USB HUB ( See Electrical Diagram in Appendix A).
Figure 4 Mapping Software Interface
3.3.3 GPS Data Collection Format
The file name of the data was named as current date and time ( military time) with extension “ txt”
(“ yyyymmdd hhmmss. txt”, 20060918 103414. txt). There is a new file every sixty minutes,
20060918 113414. txt. There is also a new file at midnight, 20060919 000000. txt. A new file is
also created when the system time is changed except when changing minute and second. For
example, there will be a new file when changing system year, month, day or hour. Changing
minute will not be effected in creation a new file unless minute is changed to the time before file
was created or sixty minute after the file was created.
The data is written as date, time latitude, longitude, heading and velocity per line, as follow,
every time a new data is updated.
13
09/ 18/ 06 23: 34: 34, N 43 0.0591, W 70 0.0104, 265.10, 14
09/ 18/ 06 23: 34: 36, N 43 0.0564, W 70 0.0104, 265.30, 14
09/ 18/ 06 23: 34: 38, N 43 0.0520, W 70 0.0205, 255.04, 16
.
.
09/ 18/ 06 23: 59: 59, N xx x. xxxx, W xx x. xxxx, xxx. xx, xx
3.3.4 GPS Data Display and Mapping
GPS interface is connected through 4- Port USB HUB ( See Appendix A) and the location data
from the GPS receiver is recorded and examined to check for their variability and consistency.
Shown below in Figure 5 is an exemplar set of GPS data, which indicates a trace of the vehicle
trajectory. Figure 6 depicts another set of GPS data traces with an overlap of the local street
maps near the Berkeley campus.
15.35
15.4
15.45
15.5
15.55
15.6
15.65
15.7
15.75
52.04 52.06 52.08 52.1 52.12 52.14 52.16 52.18 52.2
point3 N 37
52.1336 W 122
point2 N 37 52.1268
W 122 15.5315
point9 N 37
52.1758 W 122
N 37 52.1331 W 122 15.4999
N 37 52.1375 W 122
N 37 52.0912 W 122
N 37 52.0539 W 122 15.6697
N 37 52.0996 W 122 15.6855
point1 N 37 52.1525 W 122
Barrow Ln
Bancroft
W
Dana St
Durant
Bowditch St
Figure 5 GPS Data Coordinate Display
14
Figure 6 GPS Data Display Overlapping with Local Street Map
3.3.5 Summary of UCB- PD Project 54 and GPS Evaluation
The results of GPS data experiments and Project54 at UCB- PD can be summarized as follows:
( 1) Low- cost GPS devices are readily available with decent resolutions and performance
specifications for the purpose of incident reporting.
( 2) GPS Data recording in the Berkeley neighborhood, a suburban region with scattered tall
buildings and occasional narrow streets, was consistent and reliable for its intended usage.
( 3) Mapping and navigation was a function preferred by officers during the testing.
( 4) Project54 is an easy- to- use system with minimum learning curves. However, the scale of
experimentation with one single vehicle was too small to fairly assess its benefits.
( 5) The radio units in the UCB- PD cruiser were not compatible with Project54. As a result, the
overall performance was significantly reduced.
( 6) More training and user interactive sessions are strongly desirable.
15
4. PHOTOGRAMMETRY
One primary factor in the delays of incident handling is the necessary and proper documentation
of incident information for legal and technical reasons. The prevailing method of scene
measurement is often conducted by physically walking the scenes with wheel measures or
similar tools. The common method and its conventional approach present several issues: time
consumption, road hazard, limitations at site, and the needs for photographic evidence. In this
project, we investigated and developed a computer software application using photogrammetric
techniques to facilitate the measurement of the aforementioned incident scenes. This software
will allow the user to simply take one or a few photographs and extract all needed measurements
in the convenience of the office, solving the shortcomings of the conventional method.
4.1 Photogrammetry Basics
Photogrammetry as defined by the American Society for Photogrammetry and Remote Sensing
( ASPRS), in the Mapping Sciences, “ is the art, science, and technology of obtaining reliable
information about physical objects and the environment through the processes of recording,
measuring, and interpreting photographic images and patterns of electromagnetic radiant
energy and other phenomena”. Or more simply put, photogrammetry is the technique of measure
2- dimensional and 3- dimensional objects from photographs using physics, geometry and
mathematics.
For geometric analysis of crash sites, a great majority of cases will only involve the identification
of points or objects on a 2- dimenaional plane. For example, occasionally it will be necessary to
pinpoint the resting positions of a vehicle by reviewing a photograph taken after a crash. In this
case, the tire- roadway contact points are the target locations to be identified by the
photogrammetry tool. For another example, sometimes it is desirable to measure the length of a
skid mark so that the pre- crash speed of a vehicle can be estimated. In this case, points along the
skid mark will be the target points to be identified with a photogrammetry tool.
For different utilization scenarios of a photogrammetry tool, the respective levels of accuracy
requirements can be quite different. When a major accident is thoroughly investigated because
of serious injuries or fatalities involved, it is often required to obtain as detailed and accurate
documented information and scene measurements as possible. This is particularly true if the
evidence and associated analysis need to be used for litigation support. On the other hand, if the
analysis of scene data extraction is used to provide supplementary information due to the
incomplete documentation at the time of on- site measurements, then the requirements will be
much relaxed. The tool developed under this project is generally intended for the latter.
4.2 Preliminary Version of Photogrammetry Application
In a previous project [ 3], a preliminary version of the photogrammetry application was
developed in the Java programming language platform. [ 1] The particular advantage of this
platform is its motto of “ write once, run anywhere”. It can be run on any operating system
without future changes and maintenance. Furthermore, not only can it be deployed as a desktop
16
application on any platform, it can also be deployed onto any web page as a Java Applet or as a
Java Webstart Application.
To use this application, a photograph taken from the incident scene must be loaded as a digital
format. Such digital formats can be obtained directly from a digital camera itself, or scanned
from a physical photograph into any of the supported digital formats. The digital formats
supported include:
JPEG: ( Joint Photographic Experts Group) www. jpeg. org
GIF: Compuserve GIF
PNG: Portable Network Graphics
The jpeg and gif formats are the most widely used in terms of storing highly compressed lossy
( jpeg) and lossless ( gif) photographs. Should other formats be used, they can be easily converted
to these formats with any convenient graphics program such as Windows Paint, Photoshop, or
ACDSee.
Figure 7 Basic Screenshot of Photogrammetry Application
Figure 7 shows the main display or interface for the developed application. The application itself
has an intuitive and easy to use Graphical User Interface ( GUI). There are two main views of the
loaded image:
• Photo View: Shows the original loaded digital image. Located in the center of the
GUI.
• Orthographic View: Shows the result of the photogrammetric techniques; where all
the measurements are shown. Located on the right of the GUI.
17
In addition to the Photo View and the Orthographic View windows, the application also contains
various tool bars, drop- down menus on the left side and top side of the displays. For example,
the two- color ( colors seen in application but not visible on report pages) rectangular window on
the left of the GUI is used for scaling and zooming adjustments of the Photo View and the
Orthographic View.
In using the application, any of the interested point in the photograph can be selected or defined
by directly clicking onto the photograph in the Photo View, resulting in a numbered point. The
result of defining a point will produce a corresponding point in the Orthographic View. Multiple
points can be defined to outline more complex objects. See Figure 8.
The projection from the Photo View to the Orthographic View yields the real- world distances
between any two of the defined points, which is shown in the right- side window of GUI. To
extract measurements from photographs of accident scenes, the simple technique of defining the
distances between two points and the length of any straight edged object does not suffice.
Usually the projection and measurement of curves are also needed. The developed application
supports this required functionality.
Figure 8 Screenshot with Multiple Defined Points on the Photo View and Corresponding
Orthographic View
4.3 Revised and Enhanced Version of Photogrammetry Application
18
The application software developed in this current project is a continuation of previous work
described in the previous section. The objectives of the software developments for this project
include modifications in several areas:
( 1) Functionality Enhancements
• Adding options for aerial survey photographs
• Providing object naming and editing options
• Providing curve- fitting options
• Allowing output diagram download
• Including output- diagram merging, if photograph scenery allows
( 2) Calculation Procedure Revisions
• Adding the option of grid generation in photos without template
• Refining grid generation methods and techniques
• Enhancing calculation and error- checking procedures
• Adding measurement unit selection
( 3) User Interface Improvements
• Revising function selection menus and windows
• Enhancing display options and color selection
4.3.1 Revised Screen Display and User Interface
Figure 9 Basic Screenshot of Revised Photogrammetry Application
Figure 9 shows the main display or interface for the developed application. The application itself
has an intuitive and easy to use Graphical User Interface ( GUI). There are two main graphical
elements of the loaded image:
􀁺 Photo View: Shows the original loaded digital image, located in the upper left portion of the
GUI.
19
􀁺 Orthogonal View: Shows the result of the photogrammetric techniques; where the
conversion of measurements from the photograph to an orthographic view are shown.
Located in the upper right of the GUI.
In addition to the Photo View and the Orthographic View windows, the application also contains
various tool bars, drop- down menus on the left side and top side of the displays.
An image is first loaded through the Open Image command under the File menu located in the
top menu bar. Images can be exported through the Export Image. Export image will save an
image along with any additional elements made by the user to the Photo View of the GUI.
The Mode menu located in the top menu bar has options for selecting different input modes. The
modes that can be chosen are: Edit ref square, Input points, and Input curves. Edit ref square
mode is the default mode when an image is first loaded. This allows a user to draw a reference
square and make changes to the reference square. Input points mode allows the user to draw
points on the image and also to draw lines between points. Input curves mode allows the user to
draw curves between multiple points ( or lines between 2 points).
In using the application, any point in the photograph can be selected or defined by directly
clicking onto the photograph in the Photo View, resulting in a pop up box asking the user to
name the point. The result of defining a point will produce a corresponding point in the
Orthographic View.
The projection from the Photo View to the Orthographic View yields the real- world distances
between any two or more the defined points, which is shown in measurements panel in the
middle of the bottom portion of the GUI. The linear Spline curve mode can be used to find the
total distance between a series of points.
In the lower left portion of the GUI, there is an input to set the size of the grid. This can be used
to set the length of the side of the reference square. There is a text box to enter the number of
units ( should be a positive real number) and a drop down box to select the type of unit ( ie. meter).
4.3.2 Aerial Photo Option
Below the grid size input, there is an Aerial photo button. This button can be used if the
photograph is taken from an overhead aerial view where the expected grid overlay is already
orthographic to the image. When this button is clicked, a pop up menu appears that asks for the
scale ( e. g. how many meters in an inch?), the unit of measurement, and how many pixels per
inch. These inputs are then used to generate a grid overlay on the image.
4.3.3 Gird Adjustment Options
Below the Aerial photo button, there is a button that is labeled " Square NOT finalized" or
" Square IS finalized" depending on the state of the grid overlay. If the reference square is drawn,
but the grid has not been drawn yet, then clicking on " Square NOT finalized" will draw the grid
overlay and cause the text of the button to become " Square IS finalized". Clicking the button
20
again will cause the grid overlay to disappear and change the text of the button to " Square NOT
finalized". The reference square can be adjusted when the text of the button reads " Square NOT
finalized".
The Delete grid button is used to delete the grid overlay, reference square, and all other
measurements. This button should be used when the user wants to start over with a new
reference square or just to get a fresh start.
The lower right portion of the GUI has control buttons to shift the orthographic view up, down,
left, and right. This can be used to shift the orthographic view when portions of the orthographic
view do not appear in the Orthographic panel.
A method to draw a more accurate grid was implemented by allowing the user to set a
measurement as a reference for the grid. By right clicking a measurement in the measurement
panel, the user can select a measurement to be used as a reference. This allows the user to draw
a reference square without knowing the measurements of it. If the user knows the length of
another measurement on the image, then the user can input the length of that measurement and
the size of the grid will automatically be calculated based on the user input.
4.3.4 Reference Square
The photogrammetry technique requires certain control points to be defined in the photograph.
These control points are known as reference points or real- world locations. For our application
we require 4 standardized control points ( the four corners of a reference square). An example of
a reference square is given in Figure 10. Once a reference square is drawn, the program can
automatically extrapolate geometric information from the reference square to overlay a grid on
top of the reference square. See Figure 11.
Figure 10 Example of A Reference Square
21
The overlay grid lies on the same plane as the reference square. In other words, the reference
square should be drawn on top of the plane of interest. For example, in Figures 10 and 11 the
reference square is drawn on the surface of the road.
Figure 11 Image Panel after Grid Generation
4.4 Accuracy of Photogrammetry Application
The accuracy of the photogrammetric measurements relies on the accuracy of the drawing of the
reference square. If the reference square is accurately drawn, then the accuracy of the
application can be reasonably high. Test cases using a 3 foot reference square showed the
photogrammetric measurements throughout the test image to be within 10% of the actual
measurements. If the reference square is carefully drawn, usually the error is less then 5%.
Photogrammetric measurements tend to be more accurate the closer they are to the reference
square. The error tends to increase as the measurement moves away from the reference square
because each pixel of error translates to a larger error.
Figures 12 and 13 show the measurement of a tape measure that is laid out close to the reference
square. The reference square is 3 feet in length and the tape measure is laid out 10 feet. The
application calculates the length to be 10.07 feet.
22
Figure 12 Measurement of a Tape Measure
Figure 13 Application Calculating Tape Length
4.5 Future Extensions
In its current state, this software application is functional and can accomplish the essential
functions in documenting roadway incidents. However, further improvements can be made to
enhance the application software. Future developments may be pursued in these areas:
23
􀁺 An improved photogrammetry technique using mathematical principles derived from linear
algebra instead of geometry, such as by a technique illustrated in Appendix B. [ 13- 15]
􀁺 Improved GUI functionality to allow more user options and greater flexibility
24
5. SUMMARY AND CONCLUSIONS
In recent years, developments in computing and wireless technological fields have advanced
considerably. Many devices and associated software are now readily available off the shelf.
Selective use of such products will enable the implementation of driver- assistance systems for
law- enforcement applications. In this project, we evaluated the applicability of GPS, Project 54,
and Photogrammetry tools to assist law enforcement officers in handling and documenting
crashes.
5.1 Scope of Technology Survey and Experimentation
During the course of the project, the research team carried out a broad survey and assessed the
availability and suitability of individual devices and packaged systems for law enforcement
applications. After initial evaluation, a collaborative relationship was established with the
University of New Hampshire CATlab, where Project54 ™ was developed. This cooperative
effort allowed us immediate access to the mature and already deployed technology set that
appears to be an ideal candidate to be considered for CHP and state agency vehicle fleets.
The research team also collaborated with the city of Carlsbad Police Department, where a Public
Safety Technology Problem was implemented with great success. Considerable resources were
dedicated to implement infrastructure and vehicle renovations to provide officers with mobile
office capability.
In addition, the research team deployed a Project 54 system at the Police Department of the
University of California at Berkeley on an experimental vehicle on one UCB- PD cruiser.
Experiments and user surveys were carried out to explore technical and institutional issues in
using a driver- assistance system for law enforcement functions. Furthermore, work was also
conducted to evaluate the use of vehicle- mounted GPS device for the purpose of vehicle location
identification and incident reporting.
In this project, we also developed and enhanced a computer software application using
photogrammetric techniques to facilitate the measurement of the aforementioned incident scenes.
This software will allow the user to take photographs taken at the crash sites and extract
supplementary measurements in the convenience of the office, mitigating the shortcomings of
time- limited on- site documentations.
5.2 Summary of Findings
The evaluation of Project 54 provides tremendous insight into the keys to successful
implementation of vehicular- based assistance for law- enforcement officers:
( 1) Flexibility and diversity of user interfaces, which allows the users to activate on- board
devices under various field operating conditions.
( 2) Short learning curves with user- friendliness design.
( 3) Non- proprietary open architecture, which allows the flexible choice of modular components
and sub- systems.
25
Based on the feedback from the Carlsbad Police Department, their successful Public Safety
Technology program can be contributed to:
( 1) The system- wide implementation of Information Technology Architecture, and working
closely with software vendor.
( 2) Selecting the appropriate and user- friendly systems, such as Project 54.
( 3) Multiplicity of functions giving officers mobile office capabilities.
( 4) Favorable cost- benefit returns for Agency investment.
The results of GPS data experimentation at UCB- PD can be summarized as follows:
( 1) Low- cost GPS devices are readily available with decent resolutions and performance
specifications for the purpose of incident reporting.
( 2) GPS Data recording in the Berkeley neighborhood, a suburban region with scattered tall
buildings and occasional narrow streets, was consistent and reliable for its intended usage.
( 3) Mapping and navigation was an application preferred by officers.
5.3 Recommendations
The recommendations as a result of the work described in this project can be summarized as
follows:
( 1) Close collaboration and frequent interaction between the developers and the users are
essential for successful implementation of technology programs.
( 2) Positive feedback from officers is achievable and apparent benefits for local jurisdictions can
be expected, as evidenced in the New Hampshire Project54 Project and the City of Carlsbad
Public Safety Program.
( 3) Sufficient flexibility must be built into the user interfaces so that users can adopt a level of
utilization according to his/ her personal preferences.
( 4) There may be significant institutional issues involved in the deployment of technology
systems. Early participation and coordination by all interested parties is desirable.
( 5) Extensive user training and extended phase- in periods could minimize the obstacles and
mitigate the side effects.
26
REFERENCES
[ 1] C- Y. Chan, Ray Su, “ A Mobile Platform for Roadway Incident Documentation,
“ California PATH Research Report, UCB- ITS- PRR- 2004- 2, January 2004.
[ 2] C- Y. Chan, Ray Su, “ A Software Application of Photogrammetry Techniques in
Reconstructing Incident Scenes,“ California PATH Research Report, UCB- ITS- PRR- 2004-
3, January 2004.
[ 3] SWITRS, Statewide Integrated Traffic Records System, California,
http:// www. chp. ca. gov/ html/ aiuswitrs. html
[ 4] TASAS, Caltrans Traffic Accident Surveillance and Analysis System
http:// www. dot. ca. gov/ hq/ traffops/ signtech/ signdel/ chp3/ chap3. htm# 3- 04.
[ 5] http:// www. trimble. com/ gps/ index. shtml: Trimble GPS Tutorial.
[ 6] http:// www. navcen. uscg. gov/ ndgps/ default. htm: US Coast Guard Webpage.
[ 7] http:// www. tfhrc. gov/ its/ ndgps/ handgps/ 03039. htm: FHWA HA- NDGPS Webpage.
[ 8] http:// www. path. berkeley. edu/ PATH/ Research/ current/ safety/ 6217. html: California PATH
VII California.
[ 9] Evaluation of the High Accuracy- Nationwide Differential Global Position System ( HA-NDGPS)
for the California PATH Program, FINAL REPORT prepared for California
PATH, November, 2006, revised December, 2006, Bourns College of Engineering, Center
for Environmental Research and Technology, University of California, Riverside, CA
92521
[ 10] M. Barth, M. Todd, “ Differential GPS Architectures for the VII California PATH Program
UC Riverside, Meeting Presentation, March 1, 2007.
[ 11] E. R. Green et al., “ Evaluation of the Accuracy of GPS as a Method of Locating Traffic
Collisions,” Kentucky Transportation Center Report, KTC- 04- 08/ SPR- 276- 04- 1F, June
2004.
[ 12] http:// www. tsc. berkeley. edu/ html/ res_ GIS_ switrs. html: University of California at
Berkeley, Traffic Safety Center GIS Page.
[ 13] http:// www. project54. unh. edu/: CATLAB, University of New Hampshire.
[ 14] www. project54. unh. edu/ Reference/ Download. pm/ 2173/ Document. PDF, Law Officer
Magazine, September/ October 2005.
[ 15] http:// www. robots. ox. ac. uk/~ vgg/ presentations/ bmvc97/ criminispaper/ node2. html# SECTi
ON00020000000000000000
[ 16] Mundy J. and Zisserman A. Geometric Invariance in Computer Vision. MIT Press, 1992.
[ 17] Semple J. and Kneebone G. Algebraic Projective Geometry. Oxford University Press,
1979.
27
Appendix A: Experimental Vehicle Instrumentation
A. 1 Computer Installation
The Project54 system embedded computer is installed in the right side of the trunk of the vehicle
( Figure A. 1). The made and model of the computer is, Advantech, PCM 9371.
Figure A1 Equipment Mounted inside the Trunk
A. 2 Power supply Installation
Figure A2 Power Supply and Ignition- Switch Controlled Circuit Breakers
The system is powered by DC to DC ( OPUS Solution, Inc. Model: DCA. 080.512) time delay
power supply ( Figure A2) directly from the car battery through 10A circuit breaker and 60A
circuit breaker. The 10A switch/ circuit breaker ( Figure A2) is located by the computer and the
breaker is used to turn on 30A relay which provides power to 12V bus bar. The 60A circuit
breaker ( Figure A2) is located in the engine compartment. However, the power supply is ON
28
only when ignition signal is received by ignition signal ( Figure A3) input of power supply. The
power supply will be turned off approximately one minute after the ignition is turned off.
A. 3 Interface Boxes ( IDB Boxes)
There are five IDB ( Intelligent Transportation systems Data Bus) boxes installed in the trunk.
Radio and Light Bar IDE boxes are located near the radio and light bar control ( Figure A5).
Computer, Microphone and GPS IDE boxes are located by the computer ( Figure A2). All IDB
boxes are connected in series by network cable. For proper termination, the GPS box must be
last unit in the network.
Computer IDB box is directly connected to the computer serial port and powered by 12V bus bar
as shown in Figure A3.
Microphone IDB box is daisy chained to computer IDE box with network cable. The parallel
port of the Microphone IDE box is used for executing voice command function. The execution
is done by connecting the two wires of the parallel port when push to talk button is pressed
( Figure A8). Since Ford Police car is not equipped with the cruise control option, one of the
cruise control buttons can be used as push to talk button. However, the buttons on the cruise
control do not make contact two conductors, instead they send different voltage levels between 0
to 5V when pushing a button. For Example, zero volt when pressing OFF button and five volt
when pressing ON button. A cable connection is required in order to link between a button on
the steering wheel and Microphone IDE box. Therefore, a simple circuit is necessary to trigger
Microphone IDE box to listen voice command. The circuit between the Microphone IDE box
and the OFF button on the steering wheel makes contact to the two wires from Microphone IDB
parallel port by turning on relay ( Figure A3). An inverting chip 4069 is used to turn on the relay
when OFF button cruise control is pressed.
Motorola
Two- Way
Radio
Ignition
12V
Normally 5V, 0V
when the OFF
Button is
pushed.
Directional
Microphone
RJ45
GPS receiver
for mapping
5V
Whelen
Light
Bar
Control
Network Cable
data
line
adapter
GPS
Computer
Light
Bar IDB
Box
Q1
2N2222A
3
2
1
Grey
DB- 9
OPUS
Solution,
Inc.
DCA5.080.512
30A
RELAY SPDT
3
5
4
1 2
COM1
U1A
4069
2 1
10A
CIRCUIT BREAKER
1 2
PTT
RJ45
DB- 25
to Radio
Control
Head
MIcrophone
IDB Box
USB
12V In
DB- 25
RJ45
to
Battery
MIC
1A
FUSE
GPS
IDB
Box
12V
DB- 25
30A
CIRCUIT BREAKER
1 2
to Touch
Screen
Monitor
Time
Delay
Power
Supply
4 port
USB HUB
1M R
to Cruise
Control on
Steering
Wheel
5V
Radio Y
Cable
DB- 25
Network Cable
Out
Radio
Control
Head
Null
Modem
Cable
RJ45
D1
DIODE
DB- 9
D2
DIODE
Radio
IDB Box
Ignition
Signal
VGA
GPS
Ground
Touch
Screen
Monitor
computer
IDB Box
USB
K2
RELAY DPDT
3
4
5
6
8
7
1 2
R
This circuit 1K
turns ON K2
relay when OFF
Button on the
steering wheel
is pushed to
activate PTT.
Battery
Advantech
PCM- 9373
COM2
Key
Board
Blue
12V
DB- 9 RJ45
12V
DB- 9
Figure A3 Electrical Wiring Diagram
29
Radio IDB box is connected to the other end of Microphone IDE by network cable. DB25 data
port of Radio IDE box is connected to the Motorola radio and Radio control Head using “ Y”
cable. The Data Line Adapter is needed between “ Y” cable and DB25 data port of Radio IDE
box for conversion between RS- 485 to RS- 232. RS- 485 is the standard serial communication
protocol used on the Motorola radio external bus while the Radio IDE uses the RS- 232 protocol
on the DB25 port, see the diagram below ( Figure A4).
Figure A4 DB25 Port Connection
Light Bar IDB box is connected at the other end of Radio IDE network cable. Two wires ( Grey
and Blue) are used the connection between Light Bar IDE box and Whelen Light Bar control
( Figure A5) for controlling light bar.
30
Figure A5 Whelen Light Bar Control
GPS IDB box is the last box in network connection as mentioned above. The GPS receiver is
connected to the DB9 serial port of the GPS IDE box. The GPS unit is powered by computer
power supply as it is needed five volts to power. The GPS antenna is located at the top of trunk
cover ( Figure A6).
Figure A6 GPS Mounting Location
A. 4 Directional Microphone Installation
The directional microphone is installed behind the driver side sun visor. The use of a
microphone allows an officer to commands via speech to the system. Microphone is directly
connected to the system through MIC input of computer. Twelve volts power ( 12V from bus bar)
is needed to power the directional microphone. The make and model of the microphone is
Andrea, Auto Array AD- 350 ( Figure A7).
31
Figure A7 Microphone Mounting Location inside Windshield
A. 5 Monitor and Keyboard Installation
The touch screen monitor is located by the right side of steering wheel between the driver and
the passenger seat. One of the Serial ports is used for communication between computer and
touch screen monitor. Also, twelve volt from bus bar in the trunk is needed to power the monitor.
Make and model is Gvision, J1PS- DA- 4266 ( Figure A8).
Keyboard is installed right in front of the monitor and USB connection is used for the Keyboard.
Make and model is Ikey, SL- 86- 911 USB.
Figure A8 Monitor and Keyboard Mounting next to Driver Seat
32
Appendix B: Camera Model for Spatial Projection in Photogrammetry
Figure B1: ( a) Plane Camera Model: a point X on the world plane is imaged as x. Euclidean
coordinates X- Y and x- y are used for the world and image planes, respectively. Φ is the camera
centre. ( b) One- dimensional Camera Model: The camera centre is a distance f ( the focal length)
from the image line. The ray at the principal point p is perpendicular to the image line, and
intersects the world line at P, with world ordinate t. w is the angle between the world and image
lines.
Figure B1a shows the imaging process. The notation used is that points on the world plane are
represented by upper case vectors, X, and their corresponding images are represented by lower
case vectors x. Under perspective projection corresponding points are related by [ 13, 14]:
X= Hx
where H is a 3X3 homogeneous matrix, and ``='' is equality up to scale. The world and image
points are represented by homogeneous 3- vectors as X = ( X, Y, W) T and x = ( x, y, 1) T. The scale of
the matrix does not affect the equation, so only the eight degrees of freedom corresponding to the
ratio of the matrix elements are significant.
The camera model is completely specified once the matrix is determined. The matrix can be
computed from the relative positioning of the two planes and camera centre. However, it can also
be computed directly from image to world point correspondences.
33