Image Edge Detection Using Parallel Depth Learning


  • Peng Xu,Xiaohui Zhang , Zhihua Zhai


With the development of scientific information technology, image processing and
computer vision technology have gradually become an important research direction. At
the same time, deep learning has been greatly developed and become a technological
change. Various kinds of deep neural networks are constantly emerging. Therefore, the
application of deep learning methods and probing neural networks to edge detection
has become a new trend. In this paper, the traditional differential operator edge
detection method, deep learning and deep neural network theory are systematically
expounded. On this basis, the advantages and disadvantages of edge detection operator
are analyzed, and the edge detection technology based on parallel depth learning is
introduced. Through the analysis and validation of the experimental results, the parallel
depth learning edge detection technology has a strong target understanding ability, is
more suitable for extracting the contour of the target object, effectively suppresses the
edge of non-target object, and is helpful for subsequent target detection and analysis.