Products

07 July 2011

Mobile machine vision system featuring AVT GigE cameras surveys pavement condition.

Proper infrastructure and road maintenance is essential to ensure the safety of road users and to keep ve-hicle operating costs down.

Pavement assessment vehicle with front and rear camera

Pavement assessment vehicle with front and rear camera

Calibration pattern

Calibration pattern

Pavement assessment and geo-referencing in 5 steps

Pavement assessment and geo-referencing in 5 steps

AVT Prosilica GC GigE Vision camera

AVT Prosilica GC GigE Vision camera

Southeastern Surveying & Mapping Corporation (SSMC) is a land surveying company that uses mobile machine vision-based technology to perform network pavement condition assessments for the entire roadway maintained by several cities and counties in the State of Florida as well as overseas.  The company teamed up with Norpix, an Allied Vision Technologies (AVT)`s certified vision software partner, to come up with a mobile digital Geo-referencing system based on two Prosilica GC cameras from AVT and Norpix`s Streampix 5 Digital Video Recording Software.

By combining systems like Geographic Information System (GIS), GPS, and digital camera imaging, the SSMC image-based data collection system is able to collect field data and assess the condition of all roadway and pavement features such as longitudinal cracks, transverse cracks, alligator cracks, edge cracks, potholes and rutting. Image-based systems offer a less labor intensive and more reliable solution than traditional manual surveys and allow the data to be stored for future referencing.
 
125 MB/s GigE Vision® bandwidth
Two Prosilica GC1350C cameras from Allied Vision Technologies are mounted on top of a vehicle, one looking forward for an overall right of view and one at the rear of the vehicle pointing down at the pavement to capture a wide-angle view. The Prosilica GC1350 is a 1.4 Megapixel CCD camera with Gigabit Ethernet interface (GigE Vision®). The cameras are fitted with wide-angle and telephoto zoom lenses and are connected to a PC (StreamPix recording station) at the rear of the vehicle via Ethernet cabling. Both cameras run up to 10 frames per second to capture images at intervals between 5-10 feet at posted driving speeds. Up to 180,000 images a day are recorded during a surveying session. The process is facilitated by the cameras’ GigE Vision® compliant interface that is capable of a sustained maximum data rate of 125 Mbytes per second.

A GPS receiver mounted on top of the vehicle feeds global positioning data to the system, which is then linked to the image sequence and stored on the server in real-time. Captured images are displayed on screen inside the vehicle allowing the on-board technician to adjust camera settings such as exposure or gain via the Streampix 5 software. Norpix’s Streampix 5 software is a recording solution offered as turnkey system. The latest version includes features such as distance measurement information to the captured video. Systems sold also feature the latest camera models from AVT including the HD resolution Prosilica GX1920 camera with dual LAG Gigabit Ethernet interface and more.
 
Geo-referencing
The SSMC Pavement Mapping system uses a Geographic Information System (GIS) to geo-reference the captured image data to actual locations. GIS is a system designed to capture, store, manipulate, analyze, manage and present all types of geographically referenced data, and merges cartography, statistical analysis and database technology. GIS is used in many applications including land surveying, archaeology, geography, cartography, and more.

In order to geo-reference the captured image data, SSMC has developed a camera calibration grid that allows the system to associate the features on the image data to its actual location on the map. Each image is warped and
flattened and superposed on to the actual map for subsequent analysis and referencing. Images can then be enlarged and analyzed to determine the severity of the distresses. Technicians are able to identify, measure and tag distresses in the database. The stored data can easily be referenced to for pavement maintenance status updates and subsequent road network inspections months or years after the initial data was captured.

[Back to overview]