Image scene multi-object mark method based on prior condition constraint

Abstract

The invention provides an image scene multi-object mark method based on a prior condition constraint. The method comprises the steps that the region of interest of a semantic object group is determined; the multi-dimensional feature of a test image is calculated and used as a priori appearance constraint, and the pixel-level multi-dimensional feature is converted to a super-pixel-level multi-dimensional feature; the graph model structure of the region of interest of the test image is constructed; a super-pixel in the region of interest is used as a graph structure node; the adjacent relation of the super-pixel is used as the edge of the graph structure; the corresponding feature of the priori appearance constraint is converted to an edge weight value; the initial geodesic distance is calculated and used as a node weight value; geodesic propagation is carried out; in each step of propagation, the object mark of the current seed point is determined, and the geodesic distance of adjacent points around is updated to prepare for propagation of next step; and until the end of the propagation process, the object mark of each super-pixel is acquired. According to the technical scheme of the invention, the rich feature of the object is used as the priori constraint to improve the accuracy of the object mark.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a multi-object labeling method for image scenes based on prior conditional constraints. Background technique [0002] With the rapid development of social science and technology, electronic devices such as smartphones, tablet computers, and cameras are increasingly used in social life. Accompanying it is that the acquisition of image data is becoming more and more convenient, and the amount of data is also increasing. People The demand for image processing and applications is also becoming more and more abundant, and various image processing software tools are also emerging. In various industries that promote economic and social development, the demand for image scene understanding has received more and more attention. For example, in unmanned driving systems, it is necessary to understand street scenes, identify lane lines, traffic signs, and obstacle detection. , to g...

AI Technical Summary

Problems solved by technology

[0005] However, it should be noted that the semantic segmentation of objects is very difficult. It not only requires the basis of semantic recognition to distinguish between categories, but also requires the division between multiple objects within the category, and multiple objects often show various differences

At present, the work in this direction is still in an exploratory stage, and further in-depth research on theory and technology is urgently needed

Images