Image-based bus passenger number automatic statistics meter

Abstract

The invention discloses an image-based in-bus person automatic counter, comprising: image collecting and processing module and image processing algorithm, where the image collecting and processing module comprises CCD camera, video decoding chip, programmable logic chip and DSP chip, where the output of the CCD camera is connected with the input of the video decoding chip, the control and output ends of the video decoding chip are connected with the programmable logic chip, and the programmable logic chip is connected with the DSP chip; the programmable logic chip is written with a HDL program for implementing data exchange, control and buffering between the video decoding chip and the DSP chip; the DSP chip runs image processing and recognizing software. And the invention adopts high speed DSP and FPGA design and makes image recognition by advanced image processing and recognizing algorithm, and the whole system has high calculating rate and can implement real-time and accurate person number recognition; and its hardware structure is simple, having stable performance and convenient to install and debug.

Description

technical field [0001] The invention relates to an embedded digital processing circuit and an image processing algorithm, in particular to an image recognition and application technology based on a DSP digital processing circuit. Background technique [0002] The technologies currently used for bus passenger flow statistics mainly include: photoelectric detection method, pressure sensing method, photoelectric sensing method, IC card reading method and video analysis method. [0003] The photoelectric detection method requires the shape of the object to be detected to be standardized and uniform, and the motion behavior to be strictly consistent. When used in the detection of bus passenger flow, there are many factors to consider. For example: the individual differences of passengers are very large, different in height, fat and thin, and the actions of passengers getting on the bus are also different. In addition, it is possible to carry things by hand, which is easy to cause...

AI Technical Summary

Problems solved by technology

The statistical accuracy of the photoelectric sensing method can reach 92%-93% during the normal passenger flow period, but the statistical accuracy drops to 83%-85% during the peak passenger flow period, especially the statistical delay during the peak passenger flow period has a great impact on the operating efficiency of the entire system great influence

The video analysis method is complex in structure, expensive and difficult to implement

Images