Image superpixel meshing method based on fusion

Abstract

The invention discloses an image superpixel meshing method based on fusion. The method comprises the following steps of partitioning an input image by a plurality of superpixel partition methods to obtain a plurality of superpixel partition results, carrying out edge extraction, linear fusion and normalization processing in sequence, and performing distance conversion on a processed result to generate a soft edge graphic; setting mesh points by a Hungary algorithm; calculating an edge between adjacent mesh points according to an active contour to update the soft edge graphic; endowing each image region which is surrounded by four mesh points and the edges among the mesh points with a corresponding label which uniquely determines the image region, thus generating mesh superpixels. According to the method, the computational order of magnitudes is reduced, the operation speed is improved, the partitioning effect of the mesh superpixels is good on the edge of an object, the detection precision is improved, the need in the actual application is met, and according to an experiment result, the method provided by the invention can obtain more accurate detection result within shorter computation time.

Description

technical field [0001] The invention relates to the fields of multimedia information processing and computer vision, in particular to a fusion-based image superpixel gridding method, which can be used for fast and accurate image processing, identification and matching. Background technique [0002] At present, in the field of computer vision, when solving problems such as image recognition and matching, the similarity between the two regions is extracted by calculating the pixel-level features in a corresponding region of the two images, so as to perform specific Object recognition and matching. Simply put, superpixel-level image processing, recognition and matching methods are a method for detecting identical objects or similar regions between two or more images at the superpixel level. [0003] In the prior art, regional features, that is, regional descriptors, are usually extracted from all color channels of all pixels in a region of each image, and by comparing the simi...

AI Technical Summary

Problems solved by technology

[0005] First of all, for an area in the image, feature extraction is performed on all pixels, and the amount of calculation is very large;

[0006] Secondly, the existing methods can only delineate a region in the image as a rectangular region, so for objects with irregular edges, the detection results obtained are poor

Images