Reserch And Application Of Corner Extraction Of Remote Sensung Image Based On Edge Feature Detection

In recent years,China's emphasis on emerging high-tech industries has made the development of China's remote sensing science and technology even more rapid.In the development of national civil space infrastructure,and the satellite system has been gradually improved,and the acquisition and quality of remote sensing data have been continuously improved.However,it is inevitable that the remote sensing images will be disturbed by many complex environments such as the atmosphere,transmission media,features of the ground features,and camera imaging during the imaging process.Therefore,when extracting the remote sensing image information,by correspondingly processing the target image,the ability to further express the remote sensing image information can be achieved.In view of the above problems,in the study of remote sensing image feature extraction,this paper proposes a method of corner extraction of remote sensing images based on edge features.And applied this method to the image processing platform based on JGraph,realized the visualization of the image processing process.The specific research work and results are as follows:1.Realize the edge feature detection of remote sensing image based on wavelet transformAfter comparing several common differential edge detection operators to the edge detection of remote sensing images,it is found that the effect is not ideal.Either the noise in the image has too much influence,or the detection operator operation makes the edges too smooth.To solve this problem,the edge feature detection of remote sensing image based on wavelet transform is proposed.According to the characteristics of wavelet algorithm,it can decompose the target image,obtain the high-frequency and low-frequency components of the target image,and then use the improved threshold function to process the noise of the image,and achieve the purpose of image edge feature extraction.The comparison experiment shows that the improved wavelet transform is used to extract the high-frequency components of the target image,which improves the negative effects of fine texture and noise on edge feature extraction,and better achieves the edge feature extraction of remote sensing images.2.Realization of corner extraction of remote sensing images based on edge featuresBecause the corner point is a local extreme value on the edge curve of the image,it is very sensitive to the noise in the image during extraction,which leads to a large error in corner extraction.If corner detection of remote sensing images is performed on clear edge information,the quality of remote sensing image information extraction can be improved.Therefore,on the basis of detecting edge features of remote sensing images based on wavelet transform,Harris corner detection based on edge feature extraction is proposed.Through comparative experiments,corner detection is performed on the basis of extracting the edge features of the target image,which reduces the interference of noise on the corner extraction in the target image,and better achieves the corner extraction of the image.3.Visualization of image processing based on JGraphIn order to express the comparison experiment of edge feature detection algorithm and corner detection similar algorithm more intuitively,and to combine wavelet transform and Harris corner detection algorithm more quickly,in JGraph image processing platform,the image processing algorithm is packaged into its unique Components,to realize the visualization of the image processing process in the platform.Through the JGraph image processing platform,the image processing process can be described more intuitively,the experimental algorithm can be selected more conveniently,and the algorithm combination application can be realized more quickly.The experimental results show that the use of improved wavelet transform to extract the edge features of the target image improves the negative effects of fine texture and noise on the edge feature extraction.The corner detection is performed after extracting the edge features of the target image,so that the corner extraction is not disturbed by the noise in the target image to a certain extent,and the corner extraction of the remote sensing image is better completed.Finally,the JGraph image processing platform is used to realize the visualization of the image processing process,which not only can describe the specific process of the image processing process,but also can better select the appropriate algorithm to achieve the experimental purpose.