Simulation of Visual Image Enhancement Algorithm of Desertification Landscape Considering Visual Spatial Perception

In general, there are two main applications of image processing and analysis, one is for humans to directly extract features of interest from images. In the face of fierce market competition, the first method is obviously not suitable for the current market environment. Therefore, the cost-effectiveness of products can only be improved through image enhancement technology to meet the higher requirements of consumers. The subjective evaluation method is that multiple observers give quality scores to the enhanced images according to some pre-specified evaluation scales or their own experience, and then perform a weighted average of the scores given by all observers. Part of the noise is filtered out by grayscale correction, and a comprehensive saliency map is generated to roughly locate the crack area according to the preprocessing method of anisotropic diffusion filtering, so as to realize the precise location of cracks, and complete the design of the automatic detection and identification algorithm for pavement cracks. And detection done on the bottom layer of the image through feature fusion between high-level data. Since image information is obtained from human vision, if the visual characteristics are fully considered in image processing, the effect of image enhancement will be improved. According to the structure-function principle of ecology, the effectiveness of material flow and energy flow in the landscape mainly depends on the spatial combination of patches, matrix and corridors that make up the landscape, that is, the regional landscape pattern. People have started to study how to analyze images by computer, using the combination of image processing, pattern recognition and artificial intelligence technology. In conclusion, the final result of image enhancement should be in line with the visual perception preferences. This paper is from the perspective of perception, in order to obtain the optimal result of image enhancement. In this paper, the simulation results show that the proposed method can effectively locate the desertification landscape visual image enhancement algorithm, and has a high real-time recall rate, high detection accuracy and easy computation.