Recogniztion_2004_05_18

实现基于迭代反投影算法的超分辨率重建，对于没有噪声的情况有较好的效果。(Iterative Back Projection super-resolution.)

采用了Harrtraining和离子滤波算法实现了拥挤人群人头计数，非常有用。VC++和Opencv开发，采用Oepncv中的Harrtraining训练一个分类器即可运行(Human detection and tracking in high density crowds is an unsolved problem. Standard preprocessing techniques such as background modeling fail when most of the scene is in motion. We integrate human detection and tracking into a single framework, and in- troduce a conrmation-by-classication method to estimate condence in a tracked trajectory, track humans through occlusions, and eliminate false positive detections.)

gui图形界面导入多张图片并转换为偏振图(gui graphical interface to import multiple images and convert polarizing figure)

fbp算法实现CT重建，包括不同投影个数、探测器个数和噪声对重建后图像的质量的影响。(FBP CT reconstruction algorithm, including the number of different projection, and the number of detector noise on the reconstructed image quality.)

通过小波包变换进行工业系统中的故障识别，去噪(Wavelet Transform)

TERCOM（数字高程匹配）地形匹配算法，经调试可以运行(TERCOM (Digital Elevation match) terrain matching algorithm, the debugger can be run)

读取图像，转化为二值，去除毛刺噪点像素，去除孤立的前景像素，膨胀，对颗粒进行计数(particle count，)

利用curvelet波将不同聚焦图像进行分解，然后利用局部能量最大的方法将图像合成，得到清晰的图像、(Curvelet waves using different focus image decomposition, and then make use of local energy will be the largest image synthesis methods have been given clear images,)

结构张量来进行自适应正则化图像重建，单幅图像重建(Structure tensor adaptive regularized image reconstruction, a single image reconstruction)

对视频序列图像的压缩编码，采用了运动预测以增大压缩率(video image sequence coding used to forecast the movement increased compression rate)