Automatic Passenger Counting System 

​

​

The statistics equipment is based on the small, automatic and non-contact high-tech intelligent visual products of binocular stereo vision technology which mainly used in bus, bus, school bus and other gate/channel number statistics. The number of statistical devices can be counted in real time, or the number of people who stay in a closed area. The binocular camera captures the left and right images in the scene in real time and supports the function of infrared light. Toll device supports access two binocular cameras, support while detecting two doors or number of two-way access channels, support local area network (LAN) or network direct communication, support more than one counter work together. The dual camera of the statistics equipment can be lined up into a straight line to detect the wide gate passage of 120cm-1020cm. Toll device is through the detection and analysis through the binocular cameras vision under the height of the object, size, and direction, to judge the direction of movement of the object is "into" or "out", thus the object's "into" and "out" information real-time recording, support local store or transmission protocol. Binocular Stereo Vision, Binocular Stereo Vision) is an important form of machine Vision, it is based on the principle of parallax and use of imaging equipment to obtain the object to be tested from different positions, two images by calculating image position deviation between the corresponding points, to obtain the method of three-dimensional geometry information. Fusion of two eyes get image and observe the differences between them, we can obtain obvious sense of depth and establish corresponding relationship between the characteristics, the same space physics points in different image point in the image. These two images, which are different from each other, are called parallax images. Stereo vision binocular cameras than any of the plane (2 d)visual in terms of a single camera, based on the analysis of visual range relative distancebetween the object/height (depth) has higher accuracy.