Edge Detection With Chroma Components of Video Frame Based on Local Autocorrelation

Edge feature extraction is an important basis part for subsequent processing of video frame. The region with a small color difference is transformed into a luminance value for edge detection, which increases the difficulty of edge detection and reduces the detection accuracy. In order to improve the precision of edge detection in the regions with small color difference, local autocorrelation is proposed to enhance the edge feature. Edge detection is completed by using chroma information, and the detection rate is used to measure the extent of the detected edge of the ideal edge to be extracted. The input video frame is converted to YCbCr color mode, and local autocorrelation is applied to the luminance information Y and chroma information Cb for edge detection. Complementary processing is applied to the resulting image of edge detection based on local autocorrelation. In order to further improve the detection rate, supplementary processing is carried out. The standard edge images are defined for quantitative analysis and evaluation of proposed methods. Two kinds of edge correction processing based on the local autocorrelation comparison method and average method are proposed. Through experimental verification, the comparison method has a higher detection rate than the common method on the basis of local autocorrelation.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.