论文部分内容阅读
提出了一种基于小波变换的图像去歪斜方法．图像去歪斜的关键在于边缘检测和边框提取．边缘检测采用由租到细的方法，利用小波变换的多分辨率特性，得到的结果较传统的方法有更好的去噪和定位性能，并能保留必要的细节．边框提取则是用改进的H0ugh变换寻找直线，将参数空间两次
An image deskew method based on wavelet transform is proposed. The key to image skew is edge detection and border extraction. The edge detection method is from renting to fineness. By using the multi-resolution characteristics of wavelet transform, the result obtained by the method is better than the traditional method to denoise and locate performance and retain the necessary details. Border extraction is to use the improved H0ugh transform looking for a straight line, the parameter space twice