Image Sequence Classification Method Based on Mixed Graph Model

Abstract

The invention discloses an image sequence category labeling method based on a mixed graph model, which includes: performing superpixel segmentation on the image sequence, and performing feature description on the superpixels; performing nearest neighbor matching on superpixels between two consecutive frames of images; based on The adjacency relationship between the superpixels of a single frame image in the spatial domain and the matching relationship in the time domain between the superpixels of multi-frame images are used to model the global optimization model of the image sequence category labeling problem using the mixed graph model; the global optimization problem is solved using a linear method, and we get Class labels of superpixels in continuous multi-frame images; compared with previous graph models, the hybrid graph model constructed by the present invention can describe the first-order and symmetric relationship between superpixels in a single frame image, and can also describe two consecutive frames of images The high-order, asymmetric relationship between superpixels is solved by a linear method, effectively endowing each superpixel of the image sequence with a class label with better temporal consistency and higher accuracy.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to an image sequence category labeling method based on a mixed graph model. Background technique [0002] The image sequence category labeling based on the probabilistic graph model is an important research problem in the field of computer vision. It is a key supporting technology for applications such as autonomous vehicle environment perception and smart city traffic system construction. Its purpose is to assign a category to each pixel in the image sequence Label. Image sequence category labeling is a labeling problem, and the probabilistic graphical model is an effective tool for global optimization of this type of labeling problem. The key idea of ​​the labeling problem based on the probabilistic graphical model is to regard the label to be solved as a random variable, and obtain the global optimal solution of the random variable by minimizing the energy of...

AI Technical Summary

Problems solved by technology

In a sequence of images, multiple superpixels have a high-order, asymmetric relationship in the time domain, so they cannot be well described by simple edges

Hyperedges in hypergraphs can be used to describe high-order, asymmetric relationships between multiple variables, but traditional hypergraph models are solved based on spectral clustering methods, which are suitable for solving clustering problems and have not been applied to images. Class Labeling Problems for Sequences

Images