资源列表
TVdenoise
- TV去噪法 ,用途在於去除圖像中多餘的雜訊,並且達到輪廓分明的效果-TV denoising method, purpose is to remove unwanted noise in the image, and to achieve the effect of chiseled
jsgd-61
- 该代码采用matlab和C混合编写，是采用随机梯度的方法训练线性分类器，它可用于数据、图像等的多类分类情况。参考论文：Good Practice in Large-Scale Learning for Image Classification-This a Stochastic Gradient Descent algorithm used to train linear multiclass classifiers. It is biased towards large classificat
xiacaiyangPGLCM
- 程序主要实现了在下采样小波变换的基础上加入了灰度共生矩阵来实现图像纹理特征的提取- Program achieved a lower sampling based on wavelet transform GLCM added to achieve the image texture feature extraction
kaerman
- 用卡尔曼滤波器实现目标追踪，利用蒙特卡洛的方法对跟踪滤波器进行仿真分析，次数为1000次 -Kalman filter to achieve target tracking, using the Monte Carlo method of tracking filter simulation analysis, the number is 1000 times
Wav_denoise
- 对于原始图像中只含有较少的高频噪声，采用合适的小波分解洗漱阈值量化方法进行去噪-For the original image only contains less high frequency noise, the use of appropriate wash wavelet thresholding denoising method
Wavelet_transform
- 将信号用小波函数SYM5分解到第5层，用minimaxi阈值选择对系数进行处理，消除噪声-SYM5 decomposition using the wavelet function signals to the fifth floor, use minimaxi threshold choice to deal with coefficient, eliminate noise
iplab-2.8
- 开源图像处理软件，iplab version 2.8-open source image processing software iplab version 2.8
omp
- 压缩感知算法，正交匹配追踪算法，从收敛速度上看比传统算法优越很多，应用比较广泛。-Compressed sensing algorithm, orthogonal matching pursuit algorithm, from the point of view than the traditional algorithm convergence speed privileged lot, used widely.
SaperaDemo
- VC++调用sapera，实现图像匹配，图想相减，Blob分析。支持橡皮筋绘制矩形，设置匹配模板ROI。-VC++ call sapera, to achieve image matching, drawing like subtraction, Blob analysis. Support rubber band draw a rectangle, set the matching template ROI.
sift
- 一种sift提取特征算法的VC实现源代码，测试可以正确实现特征点的检测和提取-sift feature extraction codes by VC
FT_SalientRegionDetector
- 著名的Frequency-turned saliency detection 源码-The famous Frequency-turned saliency detection source
mean_shift-recognize-source
- Menshift用于图像分割识别和跟踪，代码好用简单易懂-Menshift for image segmentation identification and tracking, code easy to use easy to understand