A method for high-score image change detection based on color space smoothing and improved frequency tuning saliency model

Abstract

The invention discloses a high-score image change detection method based on color space smoothing and an improved frequency tuning salient model, comprising the following steps: obtaining an image difference map and a logarithmic ratio map, decomposing the difference map and the logarithmic ratio map, and then fusing the decomposed image to obtain a fused difference image; on the basis of the fusion difference image, the color space smoothing based on Mean Shift being applied to the fusion difference image to remove the disturbance region and retain the change region in the fusion difference image. Aiming at the smoothed fusion difference image, the improved frequency tuning saliency model is used to protect the edge of the fusion difference image, and the difference image with large variation separability is obtained. Based on the two-dimensional histogram threshold algorithm, the change information of the difference image with large change separability is extracted, and the change result is post-processed by the morphological algorithm to obtain the final change result map. The invention has high detection precision and guarantees the integrity of the detection result.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a high-resolution image change detection method based on color space smoothing and an improved frequency tuning saliency model. Background technique [0002] With the improvement of satellite image resolution, although the high-resolution remote sensing data is rich in detail information, the data volume increases and the noise interference is serious, which makes it more difficult to automatically identify changing areas in multi-temporal remote sensing images, and the corresponding data processing technology is also difficult to meet application requirements. In addition, although the edges of objects in high-resolution remote sensing images are obvious and the structure and texture information is rich, the change detection method based on feature level and pixel level cannot overcome the problem of insufficient detection accuracy at this stage; while the detection met...

AI Technical Summary

Problems solved by technology

[0002] With the improvement of satellite image resolution, although the high-resolution remote sensing data is rich in detail information, the data volume increases and the noise interference is serious, which makes it more difficult to automatically identify changing areas in multi-temporal remote sensing images, and the corresponding data processing technology is also difficult to meet application requirements

In addition, although the edges of objects in high-resolution remote sensing images are obvious and the structure and texture information is rich, the change detection method based on feature level and pixel level cannot overcome the problem of insufficient detection accuracy at this stage; while the detection method based on target level, although the accuracy is It has been improved, but there are more manual participation and a low degree of automation, and the detection accuracy depends on the target acquisition (segmentation, classification) accuracy, which affects the subsequent processing and application of change detection

[0003] Patent application number: CN201210231787.3 proposes a SAR image change detection method based on the improved C-V model. The improved C-V level set is used to segment the difference map, which has high segmentation accuracy and can effectively extract change information; However, the change information between images at different times cannot be accurately detected.

Images