This page contains a Flash digital edition of a book.
THE VIEW From Technology Creation to Deployment André Vits


How machine vision has changed the way we collect data and supports decision making


André Vits is the former Head of Unit of the European Commission’s DG INFSO andre.vits@skynet.be


W


hen Kevin Borras, editor in chief of this magazine, suggested that I might like


to write a column on machine vision, I went back through the last few issues of Tinking Highways and was surprised by the wealth of information that has been published over the last year about the subject, and indeed whatever traffic management-related topic we could discuss it seems as though machine vision has become the key technology. From the early stages of the expansion


of road traffic, engineers have made use of visual equipment to bring a sense of reality into a picture. Who doesn’t remember the pictures in the Highway Capacity Manual illustrating the definition of the Levels of Service? Te same pictures today look quite different, as traffic density and volumes have gone far beyond those levels of 1800 vehicle units per lane per hour. As video cameras came into the


market, it wasn’t long before traffic control centres were equipped with a limited number of (analogue) models. However, cost and cabling were prohibitive for wide deployment. Pneumatic tubes and inductive lopes were the standard equipment for registering traffic speeds and volumes and, naturally, manual traffic counts and vehicle classifications were carried out by students. As the first digital cameras appeared,


many researchers, including myself, saw the opportunity to process pixel data and find new ways to collect data. Major success had already been achieved in image processing, in particular in the medical sector,


68


and the availability of cheap(er) processing power that came with the PC made it possible to experiment in a real transport environment. In the early 1980s we developed a


traffic camera at the Catholic University of Leuven using a 124x64 CCD array. Te speed of the car was calculated by using two lines of the array (aſter transformation of the road surface onto the sensor) to trigger a counter, thus simulating a pneumatic speed detector. Tis resulted (aſter a lot more industrial development work) in the commercialisation by Traficon (now Flir ITS) of high performing traffic monitoring cameras. Today, we have a wide range of


cameras available that give high levels of flexibility in the design of traffic management systems at a minimum cost. But, the information provided by the manufacturers is complex and describes the functionalities from an IT point of view, than from a traffic engineering angle. Te reason is perhaps is that the final performance of the system depends not only from the sensor but on the grand design. Still, more transparency from the technology suppliers would be highly appreciated by the traffic operation managers. Perhaps the most spectacular


progress has been made in the area of automatic number plate recognition (ANPR), the core of most toll facilities, average travel speed monitoring, and enforcement schemes. For many years the automotive


industry has considered radar technology as the most appropriate for applications such as Advanced Cruise


thinkinghighways.com


Control, Lane Departure and so on. Tis technology is, however, expensive and can usually only be found in executive cars, but more camera-based systems are becoming available on the market. Te adoption of the EU Directive





We have a wide range of cameras available that give high levels of flexibility in the design of traffic


management systems at a minimum cost





2003/102, relating to the protection of pedestrians and other vulnerable road users, has triggered concerted research initiatives by the automotive industry to develop pedestrian detection systems and appropriate warning and automatic braking systems. Te Directive set limit values to be observed in the construction of the frontal structures of vehicles. Tese values should not be exceeded in a collision between a vehicle and a pedestrian. In order to ensure compliance, the vehicles would have to undergo a number of safety tests. What the Directive did do was give


the manufacturers the possibility to develop alternative measures that are at least as effective as those in the Directive. Given the difficulties and the cost associated with it and to comply with the conditions set in the directive, the manufacturers have embarked upon the development of vision-based detection (using 2D and 3D cameras). As one can understand, the underlying algorithms are quite complex due to the huge variation in shapes to be accurately detected. However, several manufacturers are offering this function in their option list and perhaps in the near future it will be standard equipment. Pedestrian detection systems are also available as an aſtermarket add-on. I believe that machine vision is to


remain one of the key points of interest in the our professional world – for both traffic management and the emerging challenges of automated driving.


Vol 8 No 2 Europe/Rest of the World


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80  |  Page 81  |  Page 82  |  Page 83  |  Page 84