Road traffic accidents are classified into three general causal categories: equipment failure, human failure, and road failure. Road-caused accidents are most often caused by a fault, damage, or degradation related to the road and its installed physical surroundings. Municipalities cannot do much to improve the risk from the first two categories, but can work to improve safety to limit road failure accidents.
Regular Inspections can Reduce Accidents
A regular or periodic road inspection can greatly help to address many road failure issues by undertaking a Paved Surface Inspection, Roadside Feature Inventory, and Roadway Asset Management. Some areas require an annual inspection of their roads. The inspections are often performed with a visual review by individuals to evaluate the surface roughness, surface distress, pavement strength, surface skid characteristics, and surface deflection. This method introduces a great deal of subjectivity and is limited by what the human operator can see and interpret. It is very time consuming, can be costly, and can produce a database with limited use.
An alternative approach is to use a camera and computer-based system that can lead to less subjectivity than human evaluation, is less costly to operate, and produces consistent and retrievable information for a reference database. Many areas in Europe that undergo regular road inspections in an attempt to find surface or road related structural failures are considering and some are even using such systems. Some places in North America are working to establish periodic inspections, but there are many opportunities to improve the efficiency and cost effectiveness of road inspections by using cameras.
An automated system for road inspection usually relies on at least five important components, though some may use more:
- the vehicle
- a location identification method, such as GPS
- the capture camera(s)
- a computing platform to capture , process and store the images
- application specific software to perform all operations such as, system to classify, store, retrieve, and interpret the data
In many cases, two computers will be required- one will be used in the vehicle to operate the capture system and to sort and store the new images and information; the second is used in a stationary mode to access previously captured images from the storage database and to provide information when needed.
- The needs for the vehicle are not particularly demanding. It is likely that a structure attached to the vehicle will be needed onto which the cameras and other equipment are mounted and aimed. Depending on the camera and computer processing capabilities and speed, the vehicle may need to be driven at a slow speed. Some systems can allow the vehicle to travel near the higher speed limits.
- The processing system is critical, but can vary based on the system demands. One important specification is the ability to process and store a very high number of images and their associated information. In addition, the accessing computer and software must quickly retrieve and access stored images and associated location information when needed. The computer that travels in the vehicle is tasked to receive and file the freshly captured images and related information. This may or may not be the same computer used to access the post-capture storage images. The computer decision depends on whether there might be a need for both situations – capturing information for road inspection and accessing the database – to be in use simultaneously, thus requiring a computer to be in both locations.
- The profiling software is unique to the application and can define the specifications and operations of the other equipment. Its primary purposes are to synchronize the captured information from different sources and to process that information. Good software packages for this application will also tie the information capture rate to the automobile speed and will allow the addition of voice or bookmark comments to the storage system during capture.
- Another requirement is to accurately track the geographic location of the vehicle when each capture event (road image) occurs and record that for reference. Most systems use a form of GPS, but other methods are possible. The location data is logged right with the image and other information. Norpix StreamPix software is often used in these applications. It is a full featured package that can be customized for roadway inspection. http://norpix.com/applications/southeasternsurveying.php. Photo courtesy of Southeastern Surveying and Mapping Corp.
- The camera(s) parameters are critical to having readable, storable and useful images. Often the vehicle tasked for this chore will use multiple cameras. At least one camera will be pointed at the road surface and, in some cases; multiple cameras will be used to capture surface images. In addition, it is not unusual for a system to take images of the assets around the road (signs, guardrails, etc.); requiring more cameras. This supplementary information is used to confirm the road location (used with the GPS readings) and also to inventory and assess the condition of the other roadside assets. The demands for this work can generally be met by either Charge Coupled Device (CCD) or Complementary Metal Oxide Semiconductor (CMOS) type cameras. Most often, camera selection will be based on the expected operating parameters of the system, including the anticipated speed of the vehicle, the lighting needs, and the use of the captured images, which may drive the image and information quality needed. Some design considerations for image capture include providing adequate and consistent light. It may be necessary to add a light directed to the road surface to prevent shadows and improve the captured image on cloudy days. Camera resolution generally can range from video quality to 4k x 4k high resolution, depending on how the road images will be used. The amount of information will impact the computer requirements, both for capture and eventual referencing.
The type of sensor is also a consideration as these can make a large impact. For example, roadway inspection is often done with line scan sensors. These are cameras using a single line of pixels. The images are captured by moving the camera along the roadway. The single line of pixels is gathered in the camera or computer and put together to form a “frame” which is then used as a normal area camera would output. The beauty of a line scan sensor is that the pixels are generally large and can be read out at a very high line rate. The large pixels have very good sensitivity which is required as the exposure time is very short. The faster the vehicle moves the shorter the exposure time will be. This necessitates bright illumination. The resolution of the final image is a function of the number of pixels in the sensor and the number of lines captured. Line scan sensors come in various widths. The higher the number of pixels leads to more resolution in the width direction. However, the higher the number of pixels the larger the image lens format must be. There is a trade-off between resolution and camera/lens costs. Many times additional cameras with fewer pixels are used since the cost of multiple lower cost camera/lens combinations can be much lower than a camera/lens combination that uses a higher number of pixels.
For example, the Imaging Solutions Group sells a 2k line scan camera using the Excelitas P-Series CCD sensor with 14 micron pixels. The LW-PE-2048-1394 camera uses the low-cost FireWire™ IEEE-1394 interface and a standard 35mm F-Mount for the lens.
While longer line scan sensors are available with more pixels such as 4096 and longer they necessitate larger format lenses. These can be very expensive. There are some new CMOS based sensors that have excellent performance. Sensors from Awaiba, Anafocus, E2V, Framos, Teledyne Dalsa are available. The available CMOS line scan sensors typically have smaller pixels and thus require more illumination. There are also a number of high speed area CMOS sensors such as those from CMOSIS that can be used in a line scan type mode/application.
The LW-AL-CMV-2000-USB3-M from the Imaging Solutions Group used the CMOSIS CMV-2000 area sensor with 2048 x 1088 pixels which are 5.5 micron in size. It can be used in a 2048 x 8 line mode that runs extremely fast at 26K frames per second which is equivalent to 208K lines per second. Obviously, with such a fast exposure time sensitivity is an issue and lighting is critical. The Imaging Solutions Group has also designed custom cameras for these types of applications.Some advanced systems add an infrared camera to provide images that can show pending problems based on heat measurements from the road. Anomalies in the road surface heat can be an early indicator of a pending problem and, if it doesn’t result in a more immediate repair project, the work should be included in a future schedule and budget.
There are number of possible camera interfaces to the local PC or capture system. FireWire™ IEEE-1394 is mentioned above. There is the option for Gigabit Ethernet and the corresponding GigE Vision™ industry standard protocol. Additionally, there is the new USB3 interface with the corresponding USB3 Vision™ protocol. CameraLink, CoaXPress, and other interfaces are also available.
The United States is just one of the countries that has identified infrastructure, including roads, as an area that has not received the proper attention or investment that it needs to stay safe. Using a labor and time saving approach with cameras could be an important method to improve effectiveness of the limited funds available for road improvement.
Done right, this approach will provide a sleek and affordable method to collect information and to accurately enter it into a reference or storage system for local roads. The storage system will be an invaluable tool to quickly assess any road surface that has been photographed and compare it to historical data.
Image Solutions Group (Fairport, NY) has experience with these types of systems and provides the unique camera arrangements that meet the demands of road inspection. ISG has developed cameras that have the speed to capture the needed images, even when the vehicle is moving fast, and has proven successful when installed correctly.
Expand Business Resources