Research on Application of Ant Colony Algorithm Based on Multi-sensor Image Fuzzy Fusion in Image Recognition

Abstract

In the current rapidly developing society, image recognition technology is currently an important means that can provide a full range of dynamic data. It plays an important role in economics, politics and other fields. Fuzzy image fusion algorithms have opened up to solve these problems in image recognition. Based on the current wide application of image recognition, this paper studies the research of multi-sensor image fuzzy fusion algorithm in the application of high-definition image recognition. Aiming at the characteristics of infrared and visible image fusion, an infrared-visible image fusion algorithm based on non-down sampling transformation and hybrid particle swarm algorithm is proposed. For low-frequency sub-images, an improved weighted average method based on area average is used for neighborhood fusion. For high-frequency sub-images, a hybrid particle swarm optimization algorithm is used to select the threshold, and the neighborhood algorithm based on average gradient selection is used for fusion. This algorithm can effectively improve the fusion effect of fused images. The experimental research results show that combining the characteristics of gradient intensity and phase consistency, the characteristic method of compressing the information while retaining important information, which is convenient for real-time processing, is used for related research on sensor image fusion.