A Video-Based Method for the Detection of Truck Axles Report uri icon



  • We will investigate the development of a low-cost, portable, autonomous monitoring system based on imaging technology developed for consumer digital cameras. As a first step, such a system would analyze an image and, using available pattern recognition algorithms, determine the number of axles, the separation distance, and the axle configuration, including the use of a Variable Load Suspension (VLS) system. This information would then be used to calculate the maximum allowable load for the specific vehicle configuration. Electronic signage could then notify the driver, downstream, of the maximum load, encouraging voluntary compliance. This information could also be recorder for later retrieval, providing a data collection mechanism that would help ITD estimate compliance rates by correlating the information with data collected at Ports of Entry and other sources.

    For this project, we will using existing and new videotapes of highway traffic as image sources. A frame grabber, attached to a personal computer, will be used to capture and produce digitized still images for software analysis. pattern recognition algorithms will be employed to identify axles and determine their configuration. Algorithms will be evaluated for robustness under varying image quality, such as resolution, lighting and noise. Once the image processing requirements have been determined, a prototype system consisting of off-the-shelf electronics will be specified and cost estimates performed.

    Task Descriptions:

    Task 1: Establish an image processing station capable of capturing images from video footage and processing these images. Obtain image processing hardware, software and suitable videotapes.

    Task 2: Identify appropriate image processing algorithms from the literature. Implement in software, evaluate performance and refine as needed.

    Task 3: Specify image processing requirements based on selected algorithms. Identify appropriate hardware platforms, electronics and optical components. Estimate total system costs.

publication date

  • October 2002



  • video detection; vehicle classification


report identifier

  • Budget Number. KLK474