Multi-target feature selection method and device for image classification and storage medium

Abstract

The invention provides a multi-target feature selection method and device for image classification and a storage medium, and the method comprises the steps: calculating the conditional entropy corresponding to each dimension of feature in a training sample containing multi-dimensional image features, and calculating the selected probability of the dimension of feature; initializing a preset number of particles by using a particle swarm optimization algorithm; calculating target function values of all particles, performing non-dominated sorting, and selecting a non-dominated solution to update the optimal position of a particle individual and the global optimal position of a particle swarm; when the current number of iterations reaches a preset condition, carrying out local search based on cross entropy, updating speed information and position information of particles in the local search step, calculating target function values of all particles, carrying out non-dominated sorting, and selecting a non-dominated solution to update the optimal position of a particle individual and the global optimal position of a particle swarm; and outputting a final solution by adopting an inflection point selection method under the condition that the number of iterations reaches a preset number of iterations threshold.

Description

technical field [0001] The present invention relates to the technical field of image classification, in particular to a multi-object feature selection method, device and storage medium for image classification. Background technique [0002] Image classification is an image processing method that distinguishes different types of objects according to the different characteristics reflected in the image information. It uses computer to carry out quantitative analysis on images, and classifies each pixel or area in the image or image into one of several categories to replace human visual interpretation. In the implementation process of image classification, it is often necessary to extract the deep image features of the image from the feature space of the image, and then remove redundant image features through feature selection (FS, Featureselection) to reduce computational complexity. Feature selection refers to selecting multiple features from the existing D features to optim...

AI Technical Summary

Problems solved by technology

[0018] In the above existing feature selection algorithm based on particle swarm optimization, the search space of decision variables increases exponentially with the number of features. When the number of particles is much lower than the number of features, the random initialization strategy leads to poor quality of the initial solution. The search ability of the random search strategy based on the global is limited, which limits the convergence speed of the algorithm

Images