IVS

直接线性变换DLT，直接由像平面坐标到地面空间坐标的直接变换，可用于量测相机，非量测相机（本人用的是 高速相机）。配合前方交会验证l系数的准确与否。以DLT得到的结果为初值，用光束法平差可以期望得到更高精度。是数字摄影测量课程作业的例子。(Direct linear transformation DLT, and the direct conversion of coordinates directly from the image plane to the ground space coordinates, the camera can be used for measuring the non-measurement camera (I use a high-speed camera). Tie the forward intersection verify l coefficient accurate or not. To the the DLT results obtained for the initial value, bundle adjustment can expect greater accuracy. Examples of digital photogrammetry course work.)

自主选择性的查看一幅图像中的部分区域的颜色直方图：允许用户在一幅图像中选择一个矩形区域，通过按住鼠标键画一个矩形，当鼠标放开是，高亮显示矩形框，同时在另一个独立的窗口中，使用画图函数画一个图表（选定区域的颜色直方图），分别用蓝，绿，红表示选中区域中每种颜色的像素数量。(Self-selective color in an image to view a partial area of ​ ​ the histogram: Allows the user to select a rectangular area in an image by holding down the mouse button to draw a rectangle when the mouse is released, the highlighted rectangle, Meanwhile in a separate window, using the drawing functions to draw a chart (selected color histogram region), respectively, with blue, green, red indicates the number of pixels in the selected area of ​ ​ each color.)

VC++Digital image processing

过特定人语音情感数据库的建立；语音情感特征提取；语音情感分类器的设计，完成了一个特定人语音情感识别的初步系统。对于单个特定人，可以识别平静、悲伤、愤怒、惊讶、高兴5种情感，除愤怒和高兴之间混淆程度相对较大之外，各类之间区分特性良好，平均分类正确率为93.7 。对于三个特定人组成的特定人群，可以识别平静、愤怒、悲伤3种情感，各类之间区分特性良好，平均分类正确率为94.4 。其中分类器采用混合高斯分布模型。(The system of speech emotion recognition)

先进行图像预处理，比如Hough变换，边沿检测，还有差影法，模板提取 ，识别，跟踪，最后是模板匹配。(First image preprocessing, such as the Hough transform, edge detection, as well as poor shadow method, template extraction, identification, tracking, and finally the template matching.)

检测视频中的运动目标以及阴影，并消除阴影，基于opencv，将RGB颜色空间根据一定的转换关系转换到HSV空间，最后消除阴影， 可以在vc6.0和vs2008上正常运行(Shadow detection algorithm based on HSV color space)

Opencv处理图像，用ostu法实现图像的二值化(Opencv image processing, using ostu method to achieve image binarization)

注意：本程序中，每点击一次“增加亮度”，图像灰度提升16；每点击一次“降低亮度”，图像灰度降低16。(Note : This procedure, every click an "increase in brightness," gray upgrade 16; Click on each of a "lower brightness," gray lower 16.)

基于DQ效应的JPEG图像真伪检测算法，通过检测图像DCT直方图的DQ效应计算图像的概率图，该算法快速、准确，能自动定位篡改区域(JPEG image detection algorithm based on the authenticity of DQ effect, calculated image by DQ effect detection probability map image DCT histogram, the algorithm is fast, accurate, and can automatically locate tampered area)

Gray Level Edge Thinning Method Abstract. An effective edge thinning algorithm is important in image segmentation and object identification since it increases the possibility of success in detecting objects in the image and saves the processing time in the further steps such as labeling and image transformation. We investigate some gray level edge thinning method that is based on the gradient magnitude within 3x3 local neighborhood. The new method has been tested with various types of images the real world. Experimental results show that this method produces more accurate images than those of the popular non-maximum suppression methods do.