论文部分内容阅读
提出了基于小波变换零交叉的图纸扫描图像自动阈值化方法，用小波变换零交叉描述不同尺度下的图像直方图细节信息，从而聚焦到直方图的任意细节，达到对阈值的精确选取，除了考虑像素点的灰度信息外，还顾及了像素点与其邻域的边缘信息。文中从理论上给出了该方法所得阈值的物理意义。实验表明，该方法同最大墙法相比具有更强的稳健性和自适应能力。
An automatic thresholding method based on zero crossing of wavelet transform is proposed. The detail information of image histogram at different scales is represented by wavelet transform zero cross to focus on any detail of the histogram, so as to select the threshold accurately. Pixel gray-scale information, but also take into account the edge of the pixel and its neighborhood information. In this paper, the physical meaning of the threshold value of this method is given theoretically. Experiments show that this method is more robust and adaptive than the maximum wall method.