The Remote Sensing Image Segmentation Combined Denoising Method Based On Wavelet Shrinkage

At present, the information extraction of remote sensing image has been gradually become one of the hot issues in the basic research of remote sensing application, and the exploration of new algorithm of remote sensing image segmentation will make the information extraction technology constantly updated.Now the research of remote sensing image segmentation has been gradually mature,some prominent achievements have been achieved, due to remote sensing image itself with its difficulty and some other reasons, remote sensing images are inevitably subject to interference and influence by the external conditions in the process of imaging or transmission, making the image itself get the noise pollution at different degrees. Therefore, this paper proposes a novel remote sensing image segmentation model, which can be applied to segment the original image as well as the remote sensing image with noise pollution, and both achieve good segmentation results.Graph cuts and active contour model have become two important methods in the field of image segmentation, some scholars also tried to unify the two schemes to improve the computational efficiency and global minimization. GCBAC model in the realization of image segmentation has achieved good effect, but for noisy images,GCBAC model can not obtain good results. Therefore, a novel remote sensing image segmentation algorithm based on wavelet shrinkage and GCBAC model is proposed in this paper. Firstly, the wavelet transform is used to decompose the image, and the image noise pollution is eliminated by utilizing the feature adaptive threshold method,and then the GCBAC model is applied to segment the de-noised image. Experimental results show that the proposed model can achieve accurate segmentation of both the original image and the noisy image, and demonstrates strong robustness to the noise.In addition, the remote sensing image segmentation with wavelet shrinkage and level set model is also proposed in this paper, this method firstly uses the wavelet transform to decompose the image, and the image noise pollution is eliminated byutilizing the feature adaptive threshold method, at last, the improved four-phase level set model is used to segment the reconstructed image after denoising, experimental results show that the algorithm can obtain better segmentation results, and also has strong inhibition effect on the noise.