loc_im_hildith

说明：  利用卷积神经网络进行特征提取，并将卷积特征可视化(Feature extraction using convolutional neural networks and visualization of convolution features)

基于地物分类真值的高光谱遥感图像分类精度评价(Accuracy uation of remote sensing image classification)

说明：  多目标跟踪，动态背景下实现多目标跟踪算法，算法优良，实时性和鲁棒性都比较好(Tracking, dynamic context to achieve multi-target tracking algorithm, algorithm quality, real-time and better robustness than)

几种图像去燥以及图像增强的滤波器示例，中值滤波，陷波滤波器、频域低通滤波，均值滤波等(Several examples of image de dryness and image enhancement filters, median filtering, notch filter, frequency domain low-pass filtering, mean filtering, etc.)

利用低通滤波器对图像进行插值放大：按放大的倍数对原图像进行插零，插零后的图片与低通滤波函数进行卷积，原先插入的零的位置自动替换为适当的插值，如此图像就已经插值放大了，图像质量还不错。(Double the number of pixels of each row and column by adding a zero value pixel between each pixel of the image.Then operate with two kinds of low-pass filters to get the enlarged image.)

说明：  自相关算法,用于图像处理方面，可以用来处理模糊的图像(Autocorrelation algorithm, for image processing, can be used to deal with blurred images.)

视频目标跟踪中的人脸检测，包括文章加程序(Video target tracking in the face detection, including procedures for the article plus)

基于轮廓来寻找流动部分，效果不是很好，但是还可以拿来用(Based on the outline to find the flow part, the effect is not very good, but also can be used with)

说明：  内含M文件 利用对数比实现，两幅多时空SAR图像变化差异图 (Logarithm figure than the implementation differences )

AGV路径图像识别：通过CCD摄像机系统路径标线的图像，后续图像处理和识别算法性能(Image Recognition path: the path through the CCD camera system marking the images, follow-up image processing and recognition algorithm performance)