Lab7_1

CVPR2013中出现的一种显著性特征检测算法，该算法用于模拟人眼视觉识别过程，提取图像中的显著性特征借后期目标对象的检测跟踪等使用，该算法(Context-Aware Saliency Detection。matlab implementation of an classical saliency detection alrogithm,which imitate our recognition process of eyes)

一个基于Tikhonov正则化超分辨率重建，可以成功运行。(A Tikhonov based regularized super resolution reconstruction can be successfully operated.)

读取一24位彩色图像，计算其以下图像特征： 1）灰度直方图 2）颜色直方图（RGB3个通道） 3）方向梯度直方图（HOG，将图像分成4*4 16个块，8个梯度方向） (Read a 24-bit color images, calculate the following image features: 1) Histogram 2) color histogram (RGB3 channels) 3) histogram of oriented gradients (HOG, the image is divided into 4* 4 16 blocks, 8 gradient directions))

用harris算法对图像进行角点检测,输出为这些点的坐标的矩阵(Harris algorithm using corner detection image, the output for these points, the coordinates of the matrix)

该文件涵盖了65个使用IDL进行图像处理的源码例子，对IDL语言以及ENVI软件的理解非常有帮助。(The document covers 65 for image processing using IDL source code examples, on IDL and ENVI software' s understanding of language is very helpful.)

说明：  基于灰度共生矩阵的纹理特征提取方法实现目标的识别(Gray-level co-occurrence matrix based texture feature extraction methods to achieve the target identification)

对图形的旋转 有不错的效果 可以试试 让图形从左到右(Rotation pattern has a good effect can try)

自然图像的梯度分布为重尾分布，超拉普拉斯分布，本代码通过概率统计画出梯度分布的曲线(The heavy-tailed distribution of gradients in natural scenes have proven effective priors for a range of problems such as denoising, deblurring and super-resolution.These distributions are well modeled by a hyper-Laplacian)

用自相关法 来实现图像匹配，非常好用，值得一试(Autocorrelation method to achieve the image matching, very easy to use, it is worth a try)

运动检测方法的实现， 这是最原始的方法实现，根据不同的应用可以对其进行相应的改进(Motion detection method realize that this is the most primitive method, according to different applications can be improved accordingly)