MARK_Dehaze

彩色位图转黑白位图代码。何东健《数字图像处理》随书光盘提供的源码。(color bitmap to black-and-white bitmap code. Dong- Jian "Digital Image Processing" With the CD source.)

SAR复数数据转换，RAW格式数据转换，雷达图像滤波，LEE-sigema滤波，Frost滤波，Gamma-MAP滤波，KUAN滤波，雷达图像几何处理，斜距到地距转换程序。(SAR complex data conversion, RAW format, data conversion, radar image filtering, LEE-sigema filter, Frost filter, Gamma-MAP filtering, KUAN filtering, radar image geometry processing, slant range to ground from the conversion process.)

Abstract—Noninvasive ultrasound imaging of carotid plaques allows for the development of plaque-image analysis methods associated with the risk of stroke. This paper presents several plaqueimage analysis methods that have been developed over the past years. The paper begins with a review of clinical methods for visual classification that have led to standardized methods for image acquisition, describes methods for image segmentation and denoizing, and provides an overview of the several texture-feature extraction and classification methods that have been applied. We provide a summary of emerging trends in 3-D imaging methods and plaque-motion analysis. Finally, we provide a discussion of the emerging trends and future directions in our concluding remarks.

对于二值图像进行labeling,求的对象个数，每个对象的像素数，中心点，长宽高矮值等(labeling for bianry image)

这篇文章是一种特征降维的方法，不过跟一般的计较离散度的方法不同，这个特征变换的准则是：尽可能地保留原始样本的熵(Kernel Entropy Component Analysis Jenssen, R Pattern Analysis and Machine Intelligence, IEEE Transactions on Volume: PP , Issue: 99 )

（1）在工具栏中点击“打开”功能出现文件打开窗口，用户选择一幅bmp图片后在指定区域居中显示该图片，若图片尺寸小于指定区域则按原始大小显示；若大于则按合适比例显示。 （2）鼠标指针在图片上停留时显示为十字形，同时在控件下方固定位置显示鼠标所在的图像x坐标和y坐标（如：50，100），以图片左上为坐标原点。 （3）在工具栏中可以点击控件实现在图片上绘制一次直线、矩形或圆等图形，在绘图过程中要求图片和图形一直可见，不能出现刷新痕迹，绘制的图形可以任意修改形状。 （4）可以任意选择图形线条的类型（虚线或直线）、颜色（三种）及粗细（三种），要求选择方式可以叠加，绘制过程要求可任意连续执行。 （5）在工具栏中点击“保存”功能出现文件保存窗口，用户可以选择保存路径来保存当前图片操作结果（包括图片和图形）。 （6）点击“关闭”功能，关闭当前图片，若操作结果尚未保存则必须弹出对话框提示用户进行保存。 ((1) in the tool bar, click on Open function has the file open window, the user selects a bmp image to center the picture in the designated area, if the image size is smaller than the specified area press the original size if more than press the appropriate scale display. (2) is shown as a cross when the mouse pointer over the image, while displaying an image x and y coordinates of the mouse cursor in a fixed position below the control (eg: 50,100) to image the upper left of the coordinate origin. (3) in the toolbar you can click controls to achieve a draw straight lines, rectangles or circles and other graphics on the picture, the requirements of pictures and graphics in the drawing process has been seen, it can not appear refreshed traces drawn graphics can be modified shape. (4) can choose the type of graph line (dashed or straight), color (three kinds) and thickness (three kinds), asked to choose the way can be superimposed, can arbitrarily draw process requires continuous exe)

IDL编程：tif图像批量layerstacking，并保留原有地图投影。使用时只需要修改输入和输出路径。(IDL programming: tif image batch layerstacking, and retain the original map projection. Only need to modify the input and output paths use.)

遗传算法与FCM算法相结合实现图像分割，分割精度较好，速度较快(A combination of genetic algorithm and FCM algorithm to achieve image segmentation, segmentation accuracy is better, faster)

对输入的图像帧感兴趣区域进行逆透视投影变换，得到俯视图。(The input of the image frame of interest in the area of the inverse perspective projection transform, get a top view. InversePerspectiveMapping)

说明：  本课题的主要内容是图像预处理,它主要从摄像头中获取人脸图像然后进行处理,以便提高定位和识别的准确率.该模块主要包含光线补偿、图像灰度化、高斯平滑、均衡直方图、图像对比度增强，图像预处理模块在整个系统中起着极其关键的作用，图像处理的好坏直接影响着后面的定位和识别工作,内有源代码和全部论文资料(this issue is the major content of image preprocessing, mainly from the camera to obtain images Face then, in order to improve the recognition and positioning accuracy. The module consists mainly of light compensation, Grayhound, Gaussian smoothing, balanced histogram, image contrast enhancement, image pre- processing module in the system plays a crucial role in image processing will have a direct impact behind the positioning and identification, within Active code and all papers information)