bw_fragile_robust_cross

说明：  傅立叶变换在图像处理中的重要作用： (A) 图像增强与图像去噪 绝大部分噪音都是图像的高频分量，通过低通滤波器来滤除高频噪声；边缘也是图像的高频分量，可以通过添加高频分量来增强原始图像的边缘； (B) 图像分割与边缘检测 提取图像高频分量 (C) 图像特征提取 形状特征：傅里叶描述子 纹理特征：直接通过傅里叶系数来计算纹理特征 其他特征：将提取的特征值进行傅里叶变换来使特征具有平移、伸缩、旋转不变性 (D ) 图像压缩 可以直接通过傅里叶系数来压缩数据；常用的离散余弦变换是傅立叶变换的实变换； (Fourier transform in the important role of image processing: (A) image enhancement and image denoising Most of all images of high frequency noise by low pass filter to filter out high frequency noise edge is the image of the high frequency components can be enhanced by adding high frequency components to the edge of the original image (B) Image segmentation and edge detection High frequency components extracted image (C) image feature extraction Shape features: Fourier descriptors Texture: directly through the Fourier coefficients to calculate the texture features Other features: The characteristic value extracted by Fourier transform to the characteristics of translation, scaling, rotation invariance (D) image compression Fourier coefficients can be directly compressed data commonly used discrete cosine transform is the Fourier transform of the real transformation 朗读显示对应的拉丁字符的拼音)

说明：  在一幅图像中对单人脸进行定位,原理是基于多模板匹配(locate face in one picture in which there is a single face，its theory base on many templates matching)

能够提取出图像I的HOG特征，返回其幅值，角度和在各个角度上的幅值。 是进行其他针对图像的HOG特征提取的前一步。(I can extract images of HOG features, return to their amplitude, angle and perspective in all of the amplitude. Is the other for image feature extraction HOG step forward.)

该程序能够实现图像的二值化，基于C语言的。实现的效果很好(The program to achieve a binary image, based on C language. Achieve a good effect)

图象经过gabor小波滤波。gabor小波与图像卷积。(Image after wavelet filtering gabor. gabor wavelet and image deconvolution.)

彩色图像纹理提取的一种算法，效果还挺好的(color image of a texture extraction algorithm, It's a good result)

对比了常见的几种去噪方法，含中值滤波，均值滤波，维纳滤波，高斯滤波，以及三种形态学滤波（一般的，改进的，多结构元素形态学滤波）(Comparison of several common Denoising containing median filtering, mean filtering, Wiener filtering, Gaussian filtering, as well as three morphological filtering (generally improved, multi-structural elements morphological filtering))

求解L1范数最小化问题的凸优化工具包，共含有6个模型的求解方法(Solving the L1-norm minimization problem of convex optimization toolkit contains a total of six methods of solving the model)

关于指纹图像处理的有关论文资料，希望各位同仁多多 指教。(Fingerprint image on paper deal with the relevant information, I hope that Tongren exhibitions.)

利用小波变换的去噪程序！和中值滤波进行对比！(The use of wavelet transform denoising procedure! And median filtering comparison!)