FFBP

3D人脸建模过程中，重采样的重要一步，缝合面片间的缝隙。(3D face modeling process, an important step in resampling, suture surfaces between the cracks.)

一种简单高效地基于压缩感知的跟踪算法。首先利用符合压缩感知RIP条件的随机感知矩对多尺度图像特征进行降维，然后在降维后的特征上采用简单的朴素贝叶斯分类器进行分类。该跟踪算法非常简单，但是实验结果很鲁棒，速度大概能到达40帧/秒(A simple and efficient tracking algorithm based on compressed sensing. Firstly, with the random sensing matrix compressed sensing RIP conditions for multi-scale image feature dimension reduction, and then use the naive Bias classifier simple classification in the feature reduction after the. The tracking algorithm is very simple, but the results are robust, speed can reach 40 frames per second)

拉普拉斯算法的实现，程序是用vc++编写的(Laplasse algorithm)

信号进行EEMD分，作为希尔伯特黄变换的基础(try to decompose signal using EEMD，which as the Hilbert Huang Transtion basis)

(cvpr)Matching Local Self-Similarities descriptor自相似描述子((cvpr)Matching Local Self-Similarities across Images and Videos)

wu-manber算法纯c实现，多模匹配算法，效率比较高，喜欢的可以下载看看，已经调试，没有问题(wu-manber algorithm)

摘　要:舌象中的裂纹是中医舌诊中的重要内容。由于拍摄条件的不同,舌象质量有较大差异,传统的阈值选取方法并不 适用。文中提出一种舌象裂纹检测的自适应阈值选择方法。该方法利用舌象的L 3 a 3 b 色彩特征及区域的分裂- 合并, 对舌象进行区域分割,自适应地选取舌中部区域的色彩值作为阈值,对整个舌象进行裂纹提取。经验证,本方案对不同的 舌象能较好地提取出裂纹,实现舌象裂纹诊断的客观化。(Tongue’ s crack is a very important part of herbalist doctor diagnosis. For the different screen conditions , there are many differ2 ences on the tongue’ s quality. It is not effective to use the t raditional threshold ext raction algorithm. In order to solve this problem , pro2 pose a new adaptive threshold algorithm. Present the segmentation of the image of tongue by using the feature on tongue in the L 3 a 3 b color space and the split- combining algorithm. Then use the threshold , which is the color value ext racted automatically in the middle of the tongue , to detect the crack of the whole tongue. The experiments show that this approach is able to distill the crack effectively to different tongue images , and achieve the objectivity of the diagnosis based on the tongue’ s crack.)

处理视频跟踪系统中的图像跟踪，在一个复杂的背景下，不受背景震动和其他因素影响，效果还不错(Video tracking system deal with the image track, in a complex background, free from background vibration and other factors, the effect of good)

给予分水岭的图像分割，包括源代码、测试图、实验报告(Given watershed image segmentation, including source code, test plans, test reports)

读取单幅envi图像格式的img图像，并将其显示出来(read img)