Threshold segmentation method based on quick iteration of two-dimensional Tsallis gray entropy

Abstract

The invention belongs to the technical field of image processing and analysis, and discloses a threshold segmentation method based on quick iteration of a two-dimensional Tsallis gray entropy. A Tsallis gray entropy threshold selection formula based on a gray-neighborhood average gray level two-dimensional histogram is given through employing intra-class gray information in an image and combiningwith the neighborhood gray information of an image target and a background, then, the optimal threshold search process is improved by adopting a rapid iterative algorithm, the calculation efficiency of the optimal threshold is effectively improved, and meanwhile, a lookup table is established by adopting a recursion mode, so that repeated calculation of related functions in the iterative process is greatly reduced, and the speed of one-time iteration of the algorithm is higher. Therefore, according to the method provided by the invention, the target can be segmented more finely, the segmentedboundary contour is more complete, the detail texture features are clearer, the operation speed is high, the consumed time is shortest, the image segmentation performance is more excellent, and the method has a wide application prospect.

Description

technical field [0001] The invention belongs to the technical field of image processing and analysis, in particular to a threshold segmentation method based on fast iteration of two-dimensional Tsallis gray entropy. Background technique [0002] Image segmentation is the technology and process of dividing the image into several specific categories and extracting the region of interest in order to separate the target from the background. In the process of automatic target detection and recognition, image segmentation is one of the important links, and the result of segmentation will directly affect the final accuracy of automatic target detection and recognition. Therefore, image segmentation has always been a research hotspot in the field of image processing and analysis. [0003] Existing image segmentation methods mainly include edge-based segmentation methods, region-growing-based segmentation methods, and threshold-based segmentation methods. Among them, the threshold-...

AI Technical Summary

Problems solved by technology

However, if the threshold space is extended to two dimensions, the computational complexity will increase to O(L 4 ) (L is the number of image gray levels), for this reason, scholars have proposed a variety of optimization algorithms, such as the fast algorithm of the maximum Shannon entropy method, particle swarm optimization algorithm, etc., but the fast algorithm still requires a large storage space, Its threshold search speed is also affected to a certain extent; the particle swarm optimization algorithm has disadvantages such as slow convergence speed and low precision in the later stage

Wu Chengmao and others used a fast iterative algorithm to improve the optimal threshold search process, narrowed the search space, and increased the running speed, but the two-dimensional Shannon entropy threshold criterion function used tends to ignore the interaction between categories during segmentation, which affects the final result. split effect

Images