Matlab-Kalman

近两年关于图像检索的重要参考文献，包括IEEE和国内CNKI上发表的顶级期刊(Image Retrieval over the past two years on the importance of references, including the IEEE and domestic CNKI top journals published)

基于蚁群算法的图像边缘检测，将蚁群算法应用于边缘检测,提出了一种图像边缘检测的新方法，具有很大的创新性，最主要是可以仿真，供本研究方向的人参考( Image Edge Detection based on ant colony algorithm)

（1）计算图像中每个像素点的LBP模式(等价模式,或者旋转不变+等价模式)。 （2）然后计算每个cell的LBP特征值直方图，然后对该直方图进行归一化处理(每个cell中，对于每个bin,h[i]/=sum,sum就是一副图像中所有等价类的个数)。 （3）最后将得到的每个cell的统计直方图进行连接成为一个特征向量，也就是整幅图的LBP纹理特征向量； 然后便可利用SVM或者其他机器学习算法进行分类识别了。((1) calculate the LBP pattern of each pixel in the image (equivalent mode, or rotation invariant + equivalent mode). (2) then the LBP eigenvalue histogram of each cell is calculated, and then the histogram is normalized (for each cell, for each bin, h[i]/=sum, sum is the number of all the equivalent classes in a pair of images). (3) finally, the statistical histogram of each cell is connected into a feature vector, that is, the LBP texture feature vector of the whole picture. Then, SVM or other machine learning algorithms can be used for classification and recognition.)

说明：  本压缩包主要包括数字图像处理——使用MATLAB分析与实现的源代码(This compression package mainly includes digital image processing - using MATLAB analysis and implementation of the source code)

N-FINDR算法，高光谱图像无先验知识下的端元提取算法，最后计算每种端元的丰度(N-FINDR algorithm for hyperspectral image without prior knowledge of endmember extraction, and calculate the abundance)

说明：  用matlab实现了FCM图像分割，分割效果较好。有说明(Matlab achieved using FCM image segmentation, segmentation better. Described)

maximally stable extremal regions matlab codes use to detect maximally stable extremal regions.

利用块稀疏的方法求解感知矩阵，在压缩感知、图像恢复方面都可以应用(Method for solving block sparse perception matrix in compressed sensing, image restoration can be applied)

基于NSCT的几种图像融合算法，并对算法的性能作了比较，该代码有详细的说明。(Several image fusion algorithm based on NSCT, and compares the performance of the algorithm, the code is described in detail.)

brovery图像融合代码，效果比较好，还有两幅测试图像，一幅是高分辨率的全彩色图像，另一幅是多光谱图像(brovery image fusion code, the results were better. There are two test images, a high-resolution full-color images, the other is a multi-spectral image)