houghlines

用小波去噪，不同尺度下的阀值去噪，去噪效果的对比(With wavelet denoising threshold denoising different scales, denoising effect contrast)

基于NCC（归一化互相关）的图像配准算法(NCC (normalized cross-correlation), image registration algorithm)

能完成钢筋数数，图像分割，直线检测，边缘检测，很好用。(Can finish the reinforcing bar counting, image segmentation, straight line detection, edge detection and works well.)

说明：  《视觉SLAM十四讲》高翔，完整版，前10章。对视觉定位，机器学习，深度学习有很大帮助(Gao Xiang, the complete version of vision slam 14 lectures, is very helpful for vision positioning, machine learning and deep learning)

daisy是08年新提出来的特征描述符算法，常用在图像匹配等领域中。文件包含算法的PDF文档和算法提出者编写的相应matlab程序，大家可以根据自己的需要来利用。(A Fast Local Descriptor for Dense Matching，2008.)

能够将图像坐标转化为世界坐标，根据图像像素修改参数(Can transform image coordinates into world coordinates, and modify parameters according to image pixels.)

自己编写的单应性矩阵的求法，按照张正友的论文进行编写的。希望有用。(I have written the homography method, according to Zhang Zhengyou papers prepared for. I hope useful.)

对有雾的图像运用大气散射模型进行处理，通用的大四散射模型与图像的饱和度有一定关系，于是利用HSV模型中的饱和度H进行去雾处理(On a foggy image processing using atmospheric scattering model, saturation senior general scattering model have a certain relationship with the image, so the use of HSV model saturation H were treated to fog)

基于LSB的音频水印嵌入提取算法，隐藏信息为一个文本文件(Extraction algorithm, the LSB-based audio watermark embedding hidden information into a text file)

相关图像算法包括大律法分割，梯度分割，分水岭分割，活动轮廓，最小生成树，相关基因算法(Great Law Segmentation, Gradient Segmentation, Watershed Segmentation, Active Contour, Minimum Spanning Tree, Related Gene Algorithms)