IDB

卷积神经网络（Convolutional Neural Network）是一个专门针对图像识别问题设计的神经网络。它模仿人类识别图像的多层过程：瞳孔摄入像素；大脑皮层某些细胞初步处理，发现形状边缘、方向；抽象判定形状（如圆形、方形）；进一步抽象判定（如判断物体是气球）。(Convolutional neural network is a neural network specially designed for image recognition. It imitates the multi-layer process of human recognition image: pupil ingestion of pixels; some cells in the cerebral cortex preliminary processing to find shape edges and directions; Abstract judgment of shape (such as circles and squares); and further Abstract judgment (such as judging objects are balloons).)

说明：  一种实现超分辨率图像处理的新算法,本算法在综合前人算法的基础上，做了很大改进。很适合图像处理。(An implementation of the new super-resolution image processing algorithms, the algorithm in integrated algorithm based on the predecessors has done a lot of improvements. Very suitable for image processing.)

模拟单缝衍射实验，可任意给定单缝的长和宽，得出单缝衍射模拟图像、光量等高线分布图、单缝衍射3D模拟图；另外附带一个关闭全屏的小程序。(Simulation of single-slit diffraction experiment, sewn to order any length and width, draw single-slit diffraction analog image, the amount of light contour maps, 3D modeling diagram single slit diffraction addition Close Full Screen comes with a small program.)

中值滤波，对3X3有优化，纯C，正对BMP格式(Median filtering, there are optimized for 3X3)

直线段检测算法EDLines C++实现(EDLines C++ Opencv lins segment detection)

说明：  基于python-opencv的图像匹配代码，包含了SIFT、SURF、AKAZE、ORB方法。43-48行修改图像路径，74-81修改匹配方法。(dev: python3.6+opencv_python==3.4.2+opencv-contrib-python==3.4.2)

说明：  用matlab编写的图像复原源代码，采用维纳滤波复原技术，可对模糊和加噪的图像实现精确复原。(Prepared using matlab source code image restoration using Wiener filter restoration technology, can be vague and processors realize accurate image noise resilience.)

图像分割关系到目标识别的精度，是目标检测的关键一步，其中提供了一些图像分割的代码，主要为otsu算法的分割方法，很实用(s related to the accuracy of target recognition, image segmentation is a critical step in target detection, provides some image segmentation code, very useful)

去除高斯噪声的有效码源，去噪能力强，使简单。(The effective removal of Gaussian noise source code, denoising capability, so that simple.)

说明：  本文以k-means算法为背景，引入信息熵相关知识，从而实现全自动分割图像。然而在利用混合高斯模型对图像进行数据分析时，会发生一定的过拟合现象，导致我们得不到预期的聚类数目。本文设计合理的合并准则，令模型简化，有效地消除过拟合现象，使得最后得到的聚类数目与预期符合。，设计合理的准则改进了图像的全自动分割方法，使得分割结果更加优化(In this paper, k-means algorithm is used as the background, and information entropy related knowledge is introduced to realize full-automatic image segmentation. However, when the Gaussian mixture model is used to analyze the image data, there will be some over-fitting phenomenon, resulting in that we cannot get the expected number of clusters. In this paper, a reasonable merging criterion is designed to simplify the model and effectively eliminate the over-fitting phenomenon, so that the final clustering number is in line with the expectation. A reasonable criterion is designed to improve the automatic image segmentation method and make the segmentation result more optimized.)