video_avg33_filter

利用halcon mfc写了一个模板匹配demo，可以基于形状匹配，形状缩放匹配，灰度匹配对图像进行匹配

基于光流算法的图像分割，能对两帧图像检测出其中运动的物体，并对其进行分割(Image segmentation based on optical flow algorithm, two frames to detect moving object, and carries on the segmentation)

K-means彩色图像分割，有实验报告，main文件中还有区域生长的算法。(K-means color image segmentation, with experimental report.)

lsb空域水印算法，dct水印算法，dwt水印算法，处理灰度水印以及彩色图像水印，同时还有字符水印嵌入音频中。分别有着三个GUI界面实现不同功能。欢迎批评指正！(lsb airspace watermarking algorithm, dct watermarking algorithm, dwt watermarking algorithm, the watermark process grayscale and color images watermarking, as well as the characters in the watermark embedded audio. Respectively GUI interface has three different functions. Welcome criticism correction!)

利用idl编写的窗口编辑源码，在envi/idl中直接打开应用(The use of the source code editor window idl written in envi/idl directly open the application)

一些常用的高光谱解混方法的matlab代码(Hyperspectral unmixing method matlab code)

sift特征提取，包括三部分，特征点提取，特征点描述，特征点匹配(sift feature extraction consists of three parts, feature extraction, feature point description, feature point matching.)

96个蛮实用的PNG格式图标 仅供编程或图像处理使用。(96 pretty practical PNG format icon super practical!)

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).)

图像压缩实例，使用函数wdcbm2()获取信号压缩阀值(the example for image compression using wdcbm2() which is a wavelet function )