RegsiterFM

大牛人的图象恢复代码．全变分离散化的方法．经典中的经典．(Daniudi people to restore the image code. BCH-wide variable separation method. Classic classic.)

说明：  利用特征提取和分类的方法,假设输入值为{大,圆,光滑}为水果的三个特征值,来代表三种不同的水果,识别苹果\梨\香蕉\桔子\菠萝水果类别. (The use of feature extraction and classification methods, assuming that the import value was (large, round, smooth fruit) for the three characteristics of value, to represent three different types of fruit, to identify Apple \ pear \ banana \ orange \ pineapple fruit category.)

一个有关边缘检测 robert sobel FFT傅里叶变换等算法的demo(a demo using VC++ that implements the algorithms of edge detection and FFT)

说明：  识别指纹端点分叉点以及奇异点，其中还包括指纹方向场和指纹梯度场的计算(It also includes the calculation of fingerprint direction field and fingerprint gradient field)

鸟瞰图校正，基于OPENCV的应用，可以把鸟瞰图校正为平面图。(Aerial view of the correction, based on the application of OPENCV, bird' s eye view can be corrected for the plan.)

说明：  最大类间差法对视网膜血管进行分割，亲测有效。(The method of maximum class difference is effective for the segmentation of retinal vessels.)

说明：  文件说明： a1:将二进制文件转为Matalb数据 并批量保存为mat格式 a2:mat数据批量转为图片(RGB三通道） a3:RGB图片批量转灰度图、批量改变图片大小 a4:动图制作 a5:用训练好的网络对图像语义分割(Document description: A1: Convert binary files to Matalb data and save them in batch as mat format A2: mat data batch to picture (RGB three channel) A3:RGB Picture Batch Grayscale, Batch Change Picture Size A4: Motion Mapping A5: Segmentation of image semantics using trained networks)

图像融合程序，及论文 图像融合程序，及论文 图像融合程序，及论文(Image fusion procedure, and the paper image fusion procedure, and the paper image fusion process, and papers)

说明：  基于灰度矩的亚像素边缘检测，零件边缘提取(Subpixel edge contour extraction)

说明：  toolbox_calib图像畸变矫正工具箱(Image distortion correction toolbox)