Pedestrian detection system and method

摘要

A pedestrian detection system of detecting whether there is a pedestrian in a scene, the pedestrian detection system includes an image-capturing module, a preprocessing module, a human detection module, an image-stitching module and a decision module. The image-capturing module is configured for generating a plurality of first detection image data according to a contrast decision result. The preprocessing module is configured for generating a plurality of first image skeleton data according to the first detection image data. The human detection module is configured for generating a plurality of second image skeleton data. The image-stitching module is configured for stitching the plurality of first detection image data to generate at least one third detection image data. The decision module is configured for generating and outputting a detection result according to the third detection image data. A pedestrian detection method is disclosed herein as well.

主权项

1. A pedestrian detection system of detecting whether there is a pedestrian in a scene, comprising:
an image-capturing module configured for generating a plurality of image data of the scene, wherein each of the image data has a distinct exposure; for generating a contrast decision result according to a histogram of one of the image data; and for assigning at least one of the image data as a plurality of first detection image data according to the contrast decision result; a preprocessing module configured for generating a plurality of first image skeleton data labeled with regions of interests according to the first detection image data; a human detection module configured for determining whether there is a human characteristic in at least one of the regions of interests of the first image skeleton data; if so, the human detection module generating a plurality of second image skeleton data labeled with regions of human characteristics; and if not, the human detection module outputting a detection result; an image-stitching module configured for stitching the plurality of first detection image data to generate at least one third detection image data; and a decision module configured for generating and outputting the detection result according to the third detection image data, wherein the image-capturing module generates the contrast decision result according to a ratio of the number of pixels within a middle brightness range to the number of all pixels in the histogram, and wherein when the ratio is less than a threshold value, the image-capturing module selects two or more image data having distinct exposures from the plurality of image data to serve as the first detection image data; and when the ratio is not less than the threshold value, the image-capturing module selects one image data having an auto exposure (EVO) from the plurality of image data to serve as the first detection image data.