论文部分内容阅读
采用Ｂ样条小波来提取层析图像的边缘轮廓，用高斯函数为平滑函数，用Ｍａｌｌａｔ快速算法进行小波分解，分解出的局部极值就是层析图像多尺度边缘。这两种算法在不同的尺度下有不同的提取边缘细节的能力和抑制噪声的能力。做到了抑制噪声和提取边缘细节之间有机的统一。结果证明这种方法比传统的基于微分算子的边缘提取方法更适合于层析图像边缘轮廓的提取
B-spline wavelet was used to extract the edge contour of the tomographic image. Using the Gaussian function as a smoothing function, the Mallat fast algorithm was used to decompose the wavelet. The local extremum was the multi-scale edge of the tomographic image. Both algorithms have different abilities to extract edge details and suppress noise at different scales. The organic unity of noise suppression and edge extraction is achieved. The results show that this method is more suitable for the extraction of edge contour of tomographic images than the traditional edge extraction method based on differential operator