Complex image Canny edge detection method based on double sparse decomposition

Abstract

The invention relates to a complex image Canny edge detection method based on double sparse decomposition. According to the method, through structured NSCT decomposition, non-stable change high-frequency component data difficult to express in an image is separated, and a matrix of the high-frequency component data is sparse. Therefore, K-SVD dictionary learning is only carried out on a high-frequency component obtained through NSCT decomposition, a specific base vector of the image can be fused into a dictionary to form complementation with NSCT decomposition, the calculation amount of the algorithm can be greatly reduced, and the efficiency of the algorithm is improved; according to the method, a double sparse dictionary learning method is adopted, and part of stably changing high and low frequency information is possibly discarded while specific feature information of an image is reserved, so that the richness of dictionary atomic data is reduced. Therefore, a K-SVD learning dictionary and a DCT dictionary are combined, the deficiency of the learning dictionary in the aspect of stably changing a base vector is made up, and the sparse representation capability of the mixed sparse dictionary is improved.

Description

technical field [0001] The invention relates to digital image processing, in particular to a complex image Canny edge detection method based on double sparse decomposition. Background technique [0002] There are many branches in the development of image processing, such as image compression, image repair, image segmentation and image recognition. In image segmentation, image recognition and other processing methods for feature extraction, edge detection technology is needed to facilitate data reduction and information purification. The present invention uses the feature extraction function of the double sparse decomposition method to preprocess the image, removes irrelevant information, and then uses the Canny operator to detect the edge, so as to improve the accuracy of the edge detection result. [0003] Compared with the existing technology: [0004] (1) Image decomposition algorithms are mainly divided into structured signal time-frequency analysis methods and adaptiv...

AI Technical Summary

Problems solved by technology

The signal structured time-frequency analysis method is faster, but this method can only process images of a certain type of features, and the algorithm has limited adaptability

The adaptive signal analysis method has better adaptability, but the complexity of the algorithm is high, which makes the algorithm efficiency low

[0005] (2) The existing Canny edge detection algorithm has limited robustness, so the credibility and accuracy of the detection results depend on the morphological simplicity of the image

These two methods have limited adaptability and are less effective in edge detection of complex texture images

Images