NOMA

光学车间加工，很好很强大的一本书。里面有不少值得学习的东西。(Optical shop testing)

chan和vese的多水平集算法，可以用2条初始水平集来演化，从而达到检测多目标的目的。(chan and the multi-level set algorithm, you can use two sets to the initial level of evolution, so as to achieve the goal of multi-purpose.)

说明：  一种基于tv模型的彩色图像修复方法，效果很好(Tv model based on color image restoration)

IDL实现modis数据的辐射校正 修改文件路径即可用(IDL achieve modis radition correction Modify the file path can be used)

说明：  利用匹配滤波器实现图形识别，包括程序和说明文档(Graphic Recognition Using Matched Filter)

先用双峰法将细胞图像进行二值化，然后进行分割和癌细胞识别(Bimodal method first cell image binarization, and segmentation and identification of cancer cells)

说明：  在google earth engine云平台对影像进行自动分类提取，需要手动输入样本及研究区(Automatic image classification and extraction on Google Earth engine cloud platform requires manual input of samples and research areas)

针对同一场景可见光与红外图像的融合问题,提出了一种新的多尺度对比度塔图像融合方法. 该方法 利用对比度金字塔数据结构得到图像的多分辨序列,采用基于视觉特性的融合算子在图像的相应各级上融 合源图像的细节,再通过金字塔逆变换重构出最终融合图像. 这种图像处理方法具自适应性,不随各自输入 图像的灰度特性而改变,同时增强了融合图像的对比度,产生了较好的视觉处理效果.(for the same scene of visible and infrared images of integration, proposed a new multi-scale tower contrast images. The use of contrast pyramid data structure Image multiresolution sequence, based on visual characteristics of the integration operator of the image corresponding levels of integration details of the source image, through the inverse pyramid transform eventually fused images. This image processing method has adaptability without their input image with the gray characteristics change, while enhancing the integration of the contrast of the image. have a better visual effect.)

人脸识别手册，在里面详细的介绍了关于人脸检测的相关知识。不过是全英文的，如果有英语比较好，时间比较多愿意翻译这本书的人请留言。(Face Recognition manual inside detail on the face detection of the relevant knowledge. But is English, if English is better, more time is willing to translate this book, please post.)

实现了遥感图像可以进行阈值分割和分水岭分割(Threshold segmentation, watershed segmentation)