SOQcode

Harris角点检测程序（内附示例图像），利用鼠标在图像上选择提取交点的区域，进行角点提取。程序语言简洁，利于初学者学习。(Harris corner detection procedures (with sample images), using the mouse in the image on the choice of extraction point of intersection of the region, for Corner Detection. Programming language is simple and conducive to learning for beginners.)

一个很好的学习图像处理的函数的例子程序，对于初级学习者来说，有很好的帮助，对于初学习者非常适合！(a good learning image processing function examples procedures for primary learners, good help, very suitable for early learners!)

利用图像二值化方法对图像进行调幅加网，本程序实现了45度角度的调幅加网算法。(Use of the image binarization method of images in AM screening, the program achieved a 45 degrees angle of AM screening algorithm.)

典型的图像分割算法，希望对大家有用，多多支持，(Typical image segmentation algorithm, useful for all of us hope, a lot of support,)

以图像拼接目标为应用背景，对图像处理配准和融合的基本原理及相关算法进行研究；利用matlab对相关配准和融合算法进行实现并优化;并将配准和融合算法应用于图像拼接中（给定某个真实场景的一组局部图像，然后对这一组图像进行拼接，最终形成一个图像拼接系统）。要求配准和融合能单独进行工作，形成独立的程序(The basic principles and algorithms of image registration and fusion are studied with the image stitching target as the application background. MATLAB is used to implement and optimize the related registration and fusion algorithms, and the registration and fusion algorithms are applied to the image mosaics (given a set of local images of a real scene, and then to this group) The image is stitching to form an image mosaic system. Registration and fusion are required to work independently and form separate programs.)

SIFT 算法的实质上是提取局部特征的算法, 该算法在尺度空间寻找极值点, 然后根据极值点来获得尺度、旋转不变量和位置等特征。(In essence, the SIFT algorithm is an algorithm to extract local features. The algorithm finds the extreme points in the scale space and then obtains the characteristics of scale, rotation invariant and position according to the extreme point.)

利用mask与Wallis算法实现影像匀光匀色(Using mask and Wallis algorithm to realize the uniform color of the image)

CNN中的手写识别LeNet5实现，用python，进行了中文注释(CNN handwritten recognition LeNet5 implementation, using python, carried out in Chinese Notes)

从原始图像产生噪声图像，使用db1执行2层小波，使用wpbmpen进行全阈值选择，采用硬阈值和软阈值处理方式,保存低频,进行图像降噪，并绘制各分解系数图像。(Generated noise image from the original image, using db1 wavelet perform two layers using wpbmpen full-threshold selection, using hard threshold and soft threshold approach, saving low frequency, image noise reduction, and draw the decomposition coefficients images.)

说明：  使用gpu、cpu并行进行sift算子计算匹配，能够在原来的基础上加速处理，但对显卡要求较高，具体环境配置使用方法可以参照mannual(SiftGPU is an implementation of SIFT [1] for GPU. SiftGPU processes pixels parallely to build Gaussian pyramids and detect DoG Keypoints. Based on GPU list generation[3], SiftGPU then uses a GPU/CPU mixed method to efficiently build compact keypoint lists. Finally keypoints are processed parallely to get their orientations and descriptors. SiftGPU is inspired by Andrea Vedaldi s sift++[2] and Sudipta N Sinha et al s GPU-SIFT[4] . Many parameters of sift++ ( for example, number of octaves, number of DOG levels, edge threshold, etc) are also available in SiftGPU. The shader programs are dynamically generated according to the parameters that user specified. SiftGPU also includes a GPU exhaustive/guided sift matcher SiftMatchGPU. It basically multiplies the descriptor matrix on GPU and find closest feature matches on GPU. Both GLSL and CUDA implementations are provided. )