Xinxiwang Small Classroom—Application of Industrial Cameras in the Transportation Industry

        As a key component of the machine vision system , the most essential function of  industrial cameras is to convert optical signals into ordered electrical signals. Choosing the right camera is also an important part of machine vision system design. Let’s briefly introduce the application of industrial cameras in the transportation industry:

1. Vehicle detection

There are roughly four methods for vehicle detection with industrial cameras. The principle of sample point detection is to select matrix-like sample points in a certain part of the lane. When the vehicle passes, the gray scale value of the sample point is different from the original road surface. If both When the statistical value of the subtraction exceeds a certain threshold, it indicates the presence of a vehicle; detection line detection is to lay out virtual detection lines composed of pixels in the direction of vertical or parallel traffic flow to facilitate the separation of the pixel depth of the road surface and the detection line . When a vehicle passes the detection line, the gray level value on the line is different from when no vehicle passes the road; if the difference in gray level value is greater than a certain threshold, it means that a vehicle has passed. These two detection methods are generally used on single road sections where vehicles run. For complex road sections, how to lay out sample points or detection lines will become a difficulty. In addition to the two, full-screen detection is a vehicle detection method that uses the full screen for processing. This detection method can obtain more information, but the amount of data to be processed is also significantly increased. Nighttime vehicle detection is a detection method specifically designed for nighttime and darker illumination environments.

2. Vehicle identification

Vehicle identification in China is generally divided into three categories: large vehicles, small vehicles and locomotives. Industrial cameras can be used to binarize the license plate number image, and identify and record the license plate number through feature matching. Through database comparison, each number can correspond to a certain vehicle type, which can be used to capture illegal vehicles and count vehicles. , vehicle type identification, origin and destination point investigation and travel time analysis, etc. License plate recognition technology can be roughly divided into three stages: pre-processing, which binarizes the image and removes noise; license plate positioning, which uses the connected component labeling method to find the connected components in the image and analyze them to determine the license plate position; For character recognition, after character segmentation is completed, the size of the structural components is set according to the text size, and finally the text characteristics are found and compared using the typological method.

3. Vehicle tracking

The recording of vehicle trajectories in continuous images is called tracking. Vehicle tracking methods under industrial camera applications mainly include pattern-based tracking, area-based tracking, and contour-based tracking.