资源列表
myfun_SVM
- 遗传算法优化支持向量机参数的人脸识别程序，能直接运行-Genetic algorithm optimization parameters of SVM face recognition program that can be run directly
Wavelet-image-processing-
- 利用小波变换原理，将其应用到图像处理中来，对图像进行去噪，压缩，融合等。-Using wavelet transform theory, its application to image processing, the image de-noising, compression, fusion.
Urenlianshibie
- 对单个的人脸图像进行识别，通过运用网格法对人脸加以定位-Single face image recognition, through the use of grid method to locate the human face
123
- 复杂背景下车牌定位算法的研究，灰度化，二值化，按颜色特征分割出车牌-Research on complex background plate positioning algorithm, graying, binarization, characterized by color segmentation of license plates
Color-image-processing
- 彩色图像处理,例1考虑生成一幅128*128的RGB图像，例2将一幅RGB图像转换到CMY空间，例3将一幅三原色图像从RGB空间转换到HSI空间 -Color image processing, for example, an RGB image is generated considering a 128* 128, Example 2 RGB image is converted into a CMY space Example 3 a tricolor image conversion
Stroke-encoding
- 行程编码(RLE）,显示原始图像和经编码解码后的图像,显示压缩比,并计算均方根误差-Stroke encoding (RLE), the original image and the display image after encoding and decoding, the compression ratio of the display, and the root mean square error
tms320dm365
- ti dm365 datasheet tms320dm365
opencv_cvcamshift_cvmeanshift
- CAMSHIFT进行多目标的跟踪，而且根据多个场景进行改进的算法。-CAMSHIFT multi-target tracking, and improved algorithm based on multiple scenarios.
Walsh
- 主要讲通过WALSH变换进行图象处理，里面带有变换前后的图象示例-Major topics of image processing by WALSH transform image inside the sample before and after transformation with
fullAvatarEditor-1.9_20140226194838
- fullAvatarEditor插件 一款用于图片上传的插件 可用于图片的裁剪、编辑等-Crop fullAvatarEditor a feature-rich plug-in picture editor for uploading images, editing, etc.
imageDenoising
- 英伟达公司的CUDA实例代码，用CUDA实现了KNN和NLM的图片降噪算法，包含了整个对应的工程文件。-Sample: Image denoising Minimum spec: SM 1.0 This sample demonstrates two adaptive image denoising techniques: KNN and NLM, based on computation of both geometric and color distance between
recursiveGaussian
- 基于CUDA加速的高斯模糊算法，包含整个工程文件及相应测试图片。-Using CUDA implement a Gaussian blur using Deriche s recursive method. The advantage of this method is that the execution time is independent of the filter width.