siftregistration

radon算法opencv实现，能够很好的实现图像的水平旋转矫正。( implementing the radon by opencv.)

说明：  讲述PS技术的抠图经典问题，帮助你快速熟悉抠图的精华！(PS about the classic problem of technology, matting, matting to help you quickly familiar with the essence!)

用CUDA并行处理方法改写的图像降噪算法，包括KNN,NLM,以及改进的NLM方法。配置环境适合VS2010+CUDA5.0,对使用CUDA的同学来说是很好的教材。(Image denoising algorithms including KNN,NLM and refined NLM run on CUDA, useful demo for who learns CUDA based parallel programming.)

利用c++实现图像转置源代码，像素变换的一种方式(use Image home to the source code, pixel transform the way)

SAR成像算法的分析； SAR成傢算法简要介绍 ；Chirp-scaling算法 ； 分布式小卫星SAR解方位多普勒模糊的点目标仿真； 分布式小卫星SAR成像的仿真参数 ； 点目标成像仿真及结果 (SAR imaging algorithm analysis SAR algorithm married briefly Chirp-scaling algorithm distributed small satellite SAR azimuth Doppler ambiguity solution point target simulation distributed small satellite SAR imaging simulation parameters Point Target Imaging Simulation and results)

利用直方图双峰法选取适当阈值对图像进行二值化处理。此法只适用于直方图为双峰状图像。(Using the Shuangfeng method to the appropriate threshold histogram of image processing two values. This method is only applicable to the histogram of image of Shuangfeng.)

读取图像，将RGB图像转化为灰度图像，显示并保存(The RGB image is transformed into gray image)

在图像的DCT系数上嵌入信息，图像为常见的512*512大小的灰度图（实验所用图像为LENA.BMP）。嵌入的信息为一二进制Logo图（实验所用logo图为tj-logo.bmp），该图大小为32*32，共1024个Bit。 1. 在LENA图中嵌入水印，输出嵌入后的水印图像。 2. 对水印图像解码，输出解码后得到的logo图，并统计错误率。 3. 使用不同的嵌入强度进行嵌入，观察其对作品直觉效果和解码效果的影响，绘制相应的曲线。 4. 使用加入噪声、改变对比度、JPEG压缩等攻击方法对水印作品进行攻击，查看解码的效果。 (Information embedded in the DCT coefficients of the image, the image of the common 512* 512 size gray-scale (experimental image LENA.BMP). Information embedded in a binary Logo Figure (used in the experiment logo picture shows the tj-logo.bmp), the plot size 32* 32, 1024 Bit. LENA Figure embedded watermark, the watermark image output embedded. (2) logo watermark image decoding, the output decoded Figure statistical error rate. (3) using a different embedding strength for embedding, observe the works intuitive effects and decoding results, and draw the corresponding curve. (4) by adding the noise, change the contrast, JPEG compression attack methods to attack, to see the effect of decoding the watermark works.)

混合高斯(Mixture of Gaussian, MOG) 背景建模算法和Codebook 背景建模算法被广泛应用于监控视频的运动目标检测问题, 但混合高斯的球体模型通常假设RGB 三个分量是独立的, Codebook 的圆柱体模型假设背景像素值在圆柱体内均匀分布且背景亮度值变化方向指向坐标原点, 这些假设使得模型对背景的描述能力下降. 本文提出了一种椭球体背景模型, 该模型克服了混合高斯球体模型和Codebook 圆柱体模型假设的局限性, 同时利用主成分分析(Principal components analysis, PCA) 方法来刻画椭球体背景模型, 提出了一种基于主成分分析的Codebook 背景建模算法. 实验表明, 本文算法不仅能够更准确地描述背景像素值在RGB 空间中的分布特征, 而且具有良好的鲁棒性.(The background modeling algorithm of mixture of Gaussian (MOG) and codebook is widely used in moving object detection of surveillance video. However, the ball model of MOG usually assumes that the three components of RGB are independent, while the cylinder model of codebook assumes that the value of background pixel is distributed uniformly within the cylinder and the changing direction of brightness points to the origin of the coordinate system. These assumptions reduce the descriptive capability for background modeling. Therefore, the paper proposes an ellipsoid- based background model, which overcomes the MOG and codebook0s limitations. By using principal component analysis to depict the ellipsoid background model, a novel PCA-based codebook background modeling algorithm is proposed. Experiments show that this algorithm can not only give more accurate description of the distribution of background pixels but also have a better robustness.)

VC和MATLAB混合编程实现快速傅立叶变换(VC and MATLAB programming Fast Fourier Transform Mixed)