psnr-snr-enl

说明：  基于最小变分方法的快速图像复原方法，图像复原质量不变的情况下，加速算法的效率。(Image restoration based on Total Variation.)

说明：  利用matlab实现车牌定位识别，基于颜色信息，效果很明显，准确率很好(License plate location using matlab to achieve recognition, color information based on the results it is clear that very good accuracy)

说明：  语义分割，训练样本，模型训练，成果展示，深度卷积神经网络(Semantic segmentation, training samples, model training, achievement display, deep convolution neural network)

对图像分块,计算每个块的均值和方差,并给出了结果的图象化表示,易于观察(Image sub-blocks, each block calculated the mean and variance, and gives the results of the image that is easy to observe)

说明：  自己写的Active Shape Model face segmentation 代码 (Write their own Active Shape Model face segmentation code)

image processing application

一个基于Gabor滤波器组的多特征融合的图像检索算法（CBIR based on Multi-Features Combining Multi-parameters from Gabor Filter-bank, 简称multi法），融合了基于Gabor 小波系数的标准方差，峰态，能量，平滑度和一致性度量五个特征量度，实验表明，该算法是一种更加有效的图像检索方法。(we propose a CBIR algorithm based on Multi-Features Combining Multi-parameters from Gabor Filter-bank(called multi algorithm for short), which use Gabor filter banks and combine five features including standard deviation, kurtosis, energy, smoothness and uniformity together. The results of experiments show that this algorithm is a more effective method used for image retrieval.)

Adaptive Perona–Malik Model Based on the Variable Exponent for Image DenoisingThis paper introduces a class of adaptivePerona–Malik (PM) diffusion, which combines the PM

画图源程序-图像数据显示，可以对数据进行时间轴的曲线显示，也可以是空间轴的多维显示 (Paint source- the image data, the data can be curve shows the time axis, it can be multi-dimensional display space axis)

NSCT的分解与重构，用于图像处理，里面为非下采样轮廓波的工具箱(NSCT decomposition and reconstruction for image processing, which is a toolkit for subsampled contourlet)