IMAGE MATCHING METHOD BASED ON GEOMETRIC MOMENT FEATURE COUPLED WITH TEXTURE MEASUREMENT
In order to improve the matching accuracy of feature points under the variety of geometric contents,based on the introduction of bilateral filtering,this paper designs an image matching algorithm using image geometric moment feature and texture measurement.A bilateral filter was introduced to preprocess the image and filter out the noise contained in the image.Hessian operator was used to detect the image features.By using the gray value of the pixels,the geometric moments in the image feature circle were calculated to form a robust feature vector.The average gradient model was used to measure the texture features in the neighborhood of feature points,and the texture measurement results were combined with the Euclidean measurement results to accurately match the image features.With the help of cross-correlation model,the correlation degree of matching features was measured to detect the matching correctness of feature points and optimize the matching results.Experimental data show that the proposed method has higher matching accuracy and robustness than the current feature matching method under the change of various geometric contents.
Image matchingHessian operatorBilateral filteringGeometric moment featureTexture measurementAverage gradient model
万方数据
维普
