Applications of Remote Sensing Technologies for Transportation Data Collection 2004 Workshop Presentations: Questions & Answers
Road Condition Mapping using Imaging Spectrometry by M. Herold (UCSB), D. Roberts (UCSB), O. Smadi (CTRE Iowa State) and V. Noronha (UCSB) – presented by Dar Roberts, University of California at Santa Barbara
Question: Is California using this for pavement assessment actively?
Answer: No, use traditional methods (i.e. riding in car to detect bad areas).
Question: Between Digital Videography and Image Spectometry which is the more useful technology?
Answer: Near infrared can show age of the pavement, and Digital Videography has fine spatial resolution. Digital Videography has more practical uses.
Impact of Instant Imagery Access on A Regional Database for Transportation Planning - Brenda Burroughs, ORBIMAGE, Inc.
Question: Can you have historical imagery in this application?
Answer: Yes, this can be done. Historical imagery is usually available.
Question: How quick is the download of the imagery in this application?
Answer: It may depend on your machine. It is downloaded into a Bentley Microstation from the Internet.
Question: How much of this imagery can be updated?
Answer: This can be done monthly, or every 6 months.
Change Detection for Major Transportation Projects – Boston’s Central Artery – Kathleen Hancock, Virginia Tech
Question: How did you do your change analysis on low-rise to high-rise buildings?
Answer: Through visual interpretation.
Question: With this change detection analysis, how does this influence planning? If an area becomes a high-rent area, is this good or bad for the community?
Answer: Work and research will continue on and the results will have an influence on policy change.
Question: On building change analysis, did you see how many 3-story warehouses were converted to a hotel?
Answer: Not in this analysis, but you could look at tax information.
High Resolution Ortho and Oblique Aerial Imagery for Transportation GIS Database Creation and 9-1-1 Call Center Support – Art Kalinski, Atlanta Regional Council
Question: How can measurements be accurate if the pictures are oblique?
Answer: A behind-the-scenes formula is being used with trigonometry that I don’t completely understand.
Question: I thought there was a survey law in Florida and Pictometry cannot operate in the state?
Answer: This is certainly not as accurate as a orthorectified image, but Sanborn Mapping is working with them to create one large orthorectified image for an area.
Comment: This sounds similar to using stereoimagery.
How Aerial and Satellite Remote Sensing Assist Transportation Planning, Promote Environmental Understanding and Mitigate Risk – Stan Morain, University of New Mexico
Question: Does this study show a relationship between geological subsistence and population growth?
Answer: No, but it is intuitive that there is a relationship.
Transportation Centerlines: Case Studies from California and Iowa – Val Noronha, University of California, Santa Barbara
Question: Do you see the accuracy of GPS changing over time?
Answer: You have to look at price and performance. At the consumer-level, it has gotten worse. However, the software will compensate for inaccuracy.
Question: When talking about GPS versus DMI, are any DOTs replacing DMIs with GPS?
Answer: There are some DOT pilot projects, but SLDs won’t be going away.
Comment: You can get the X, Y coordinates but they are not meaningful whereas linear references are meaningful. So any GPS coordinates will need to be translated into a linear reference. GPS is being used to find a generalized location for advertising signs. Hopefully in the future X, Y and linear information can coexist.
Critical Infrastructure Issues in Emergency Evacuation of Small Areas – Rick Church, University of California, Santa Barbara
Comment: This study analyzes an existing neighborhood, but evacuation routes should be planned out before housing development is constructed.
Multi-Platform Remote Sensing for Roadway Feature Database Development - Demin Xiong, Oak Ridge National Laboratories
Question: Can you accurately interpret the results from this method?
Answer: Yes, but of course it depends on what kind of decision you need to make.
Open Forum - “New Directions for Research and Applications in Remote Sensing for Transportation Data Collection”
QUESTION from Pat Hu: "What kind of tangible benefits can remote sensing provide? What are the costs and benefits?"
(VAL NORONHA) A study conducted in Iowa showed the cost/benefit of using LIDAR in a new highway corridor. This study categorized LIDAR as having a 30-40% benefit through LIDAR as opposed to photogrammetry. Time efficiency is actually considered more important in some cases as opposed to the monetary savings for the DOT. The order of improvement with LIDAR is about 50% as opposed to the 30% improvement with photogrammetry. However, with LIDAR, the resolution is not as good. It was recommended that 2 different sets of LIDAR applications over the whole area corridor, and use photogrammetry in a limited area. This method provided a 30% improvement in cost, and a 40% improvement in time efficiency compared to traditional methods.
(STAN MORAIN) The Southwest United States was the focus of our cost/benefit study where we can’t afford as much compared to Florida. This study showed the comparative cost between LIDAR, DOQQs, and photogrammetry. Results showed faster results with LIDAR but costs more. We also conducted a cost/benefit study between IRS and IKONOS satellite images, and DOQQs. The IRS 5-m data was the best in terms of time and cost, but a specific study would need to be done for Florida.
(RICK CHURCH) There may be numerous benefits of new data types, like Pictometry. It is a simple application but invaluable in that it perhaps can eliminate crews or deploy them better. A better response time may far outweigh any monetary cost.
(DEMIN XIONG) A study conducted with Washington State DOT characterized the benefits and difficulties of remote sensing. Once you have the data, you need to ask what information can be provided for even hypothetical, subjective information, and in what specific format. We need to assess requirements, as well as what kind of product potential and the cost of its use.
COMMENT: District 3 with the FDOT prepared a cost/benefit analysis, which compared the traditional RCI collection versus photogrammetry. It was concluded that remote sensing was not a cost effective approach unless there are multiple users and uses than the its value. Please comment on the sharing of data.
(STAN MORAIN) I agree with those comments. The difficulties actually comes from industry itself. Digital Globe can share with state government but not with universities. Space Imaging can’t legally have their data shared. The aerial imagery industry may be more relaxed. We develop remote sensing applications, but don’t know what it costs at present. I hate to see a good application languish just because of cost/benefit. The first question shouldn’t be money but work environment. There is a difficulty in getting applications off the ground.
(VAL NORONHA) A microwave oven used to cost $300. Did people really think about cost comparisons on how much gas is used for the oven versus the cost of the microwave? This is a nickel & dime approach. The microwave warms the milk more quickly, so you have a happy baby. If you have an accurate centerline measurement for an emergency, then you’re dealing with the value of life. You need to worry about quality. Look at the cost of data sharing and the benefits to others. Data deals from the vendor looks like a scatter gram rather than a smooth curve.
(RICK CHURCH) If you look at the NASA program in terms of cost/benefit, it is very negative but there are plenty of spillover benefits. Computer designs and products arise out of remote sensing allot more today with transportation. Over the long run, remote sensing will play a substantial role in modeling data. For difficult models, we need better data collection. For Art Kalinski, digital oblique photos provide more information and also cost less per county. Costs will probably decrease. The playing field is changing.
(DEMIN XIONG) On one hand, there needs to cost/benefit comparison but on the other hand, you also need to make projection of future benefits. DOTs need justification for the money. There needs to be an evaluation of the perceived potential to quantify the application benefits.
QUESTION by Robert Hanson: What will be the future research in remote sensing? How do State DOTs really start using this technology?
(DEMIN XIONG) Whether remote sensing applications are to be used is a decision the DOT needs make on their own. There is a second-level potential in applications if they are broad-based in that they can be transformed by the users. To move to the application level we need to facilitate data integration to improve the whole process.
(STAN MORAIN) This is another complex question. The RSPA Consortia in Transportation recognized that the industry needed streamlining. For process streamlining, we needed research and then application development and testing. The next step was deployment, but we didn’t get there. There needs to be deployment across all of the states, so it needs to be a broad-base application that all states could use. Using remote sensing data into the ArcGIS architecture is making progress, but it is not easy. There should be an emphasis on integration to assimilate remote sensing data. I predict remote sensing in transportation will take off in 10 years.
(VAL NORONHA) There have been some exciting developments: Vehicles with videologs/ GPS; the amount of data to obtain; specialized sensors to large number of sensors. There will be real-time images to deal with future problems. In Asia, especially India, cities have a growing middle class, with an overflow of infrastructure so there is little to no parking. We can see this scenario coming in Los Angeles. Endorsed by AASHTO, we have been communicating with users, and our annual outreach to DOT users has improved greatly. We need to express ideas in meaningful ways, and double the communication efforts. Consultants will play a stronger role. We need risk takers.
(RICK CHURCH) GIS/ Remote Sensing development in Europe and North America will need allot more investment with better data, better models. After 9/11, we were worried about security. With transportation to Yucca Mountain with spent fuel rods, there is a need for more freight security. With the new DMV nominee in California, there will be no gas tax but a user tax. There is a corridor in Santa Barbara that is completely surveyed and videotaped. Remote Sensing will play niche roles, great roles, and the great integrator being GIS. Integration needs to take place to reduce the barriers, so that remote sensing plays a role.
(STAN MORAIN) I was happy to hear that the Department of Labor predicts jobs for the future will be in geospatial technology, nanotechnology, and bioinformatics. Remote sensing is involved in all, so we all have a great future being involved with this technology.
(DEMIN XIONG) RCI focuses on highways, and now there will be work on intermodal, railways, air transport, water. We will need good data. There are privacy issues, so that intermodal data on national level is getting hard to obtain.
SUMMARY
(RICK CHURCH) Much more work needs to be done in remote sensing, such as with cost/benefit studies. Completing the approaching Transportation Bill is very important so that this research can be funded.
(VAL NORONHA) We need communication with users. The National Consortia on Remote Sensing in Transportation has over 20 projects on a variety of highway infrastructure data gathering projects. Please look at the Consortia’s website (http://www.ncgia.ucsb.edu/ncrst/).
(STAN MORAIN) Ditto
(DEMIN XIONG) I would like to express appreciation to the USDOT and FDOT.