ROIDraw

自编的用于图像特征提取的源程序,解压后直接运行就ok了(writing for Feature Extraction source, unpacked directly ok on the run)

该程序实现了harris角点的提取以及其在两幅影像上的配准结果，希望对大家有用(The program implements the harris corner extraction as well as their two images on the registration results, we hope to be useful)

卷积神经网络（Convolutional Neural Network）是一个专门针对图像识别问题设计的神经网络。它模仿人类识别图像的多层过程：瞳孔摄入像素；大脑皮层某些细胞初步处理，发现形状边缘、方向；抽象判定形状（如圆形、方形）；进一步抽象判定（如判断物体是气球）。(Convolutional neural network is a neural network specially designed for image recognition. It imitates the multi-layer process of human recognition image: pupil ingestion of pixels; some cells in the cerebral cortex preliminary processing to find shape edges and directions; Abstract judgment of shape (such as circles and squares); and further Abstract judgment (such as judging objects are balloons).)

压缩感知CS——采用小波变换进行稀疏表示，高斯随机矩阵为观测矩阵，重构算法为OMP算法，对256*256的lena图处理，比较原图和OMP算法在不同采样比例(0.74、0.5、0.3)下的重构效果，并各运行50次，比较算法性能PSNR和每次的运行时间 (Compressed sensing CS- using wavelet transform as sparse representation, Gaussian random matrix as the observation matrix and OMP algorithm as the reconstruction algorithm. Compare lena figure and the reconstruction results using OMP algorithm at different sampling ratio (0.74,0.50.3),then each runs 50 times, compare the performance of PSNR and each running time)

C语言实现bmp图像放大缩小，24位bmp图像(The C language bmp image zoom, 24 bmp image)

说明：  pso 粒子群优化算法演示程序，实现图形化界面。(PSO PSO algorithm demonstration program to achieve graphical interface.)

2009年CVPR最佳论文《Single image haze removal using dark channel prior》暗原色图像去雾的算法实现。与原文不同的是本程序没有完成最后matting细化效果的部分，但仍能够达到视觉上可以接受的去雾效果。计算机视觉课程作业。工程运行于VS2008环境，需要OpenCV支持。Debug目下exe文件可以直接双击运行查看结果。(CVPR 2009 Best Paper " Single image haze removal using dark channel prior" to the dark fog of the primary image algorithm. The difference is that with the original program did not complete the final matting refinement part，yet haze-removal result is visually acceptable. Visual course work. This project runs under VS2008 with OpenCV. Double click the exe file under Debug folder to check the final result.)

水下图像的增强及边缘检测，效果很好，具有很好的参考作用(Underwater Image Enhancement and Edge Detection)

matlab下图像处理技术在智能交通中的应用 主要是小车的跟踪和标示(matlab image processing technology under the Intelligent Transportation System in the main car tracking and labeling)

本文提出了基于二维混沌映射的数字图像水印算法，混沌具有随机性、似噪声及对初始条件的极端敏感性等特点。将经过二维混沌映射置乱后的数字水印信号嵌入图像小波域的低频系数，实现了数字水印的隐蔽性、保密性和稳固性；利用二维混沌映射Arnold变换对水印信号进行置乱，不仅增强了水印信号保密性，同时有效提高了视觉上抵抗图像剪切攻击的能力。(In this paper, two-dimensional chaotic map based on digital image watermarking algorithm, Chaos with random noise and the likelihood of extreme sensitivity to initial conditions and so on. Two-dimensional chaotic map will be scrambling after the digital watermark embedded image signal of the low-frequency coefficients in wavelet domain, the realization of a digital watermark hidden, security and stability the use of two-dimensional chaotic map on Arnold transform scrambling watermark signal, not only enhances the confidentiality of the watermark signal at the same time improve the visual image shear resistance attacks.)