edgelink

图像纹理提取gabor小波提取特征一幅输入的图形进行特征提取基于gabor小波变换(image texture extraction of edge detection)

由rgb转化到ycrcb空间，再又ycrcb空间转化到rgb空间中(make a color space transmission from rgb into ycbcr，then transmit color space from ycrcb into rgb)

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)

实现遥感图像的高斯白噪声去噪，出现不同模板的去噪效果。(Gaussian white noise, noise removal, remote sensing images, different templates denoising)

AutoCAD二次开发的图像处理代码,对彩色图像进行区域生长,然后提取边界,实现图像分割.(AutoCAD second development of image processing code, color images of regional growth, and then extract the boundary, realize image segmentation.)

robustly fits a homography to a set of putatively matched image points(robustly fits a homography to a set of putat ively matched image points)

楔波变换，在SAR图像中实现边缘检测，为目标识别做铺垫工作。(The wedge waves transform, SAR image edge detection, target identification pave the way for the work.)

zernike不变矩。特征提取方法。基于形状的特征提取方法。(zernike moment invariants. Feature extraction methods. Shape-based feature extraction method.)

利用灰度积分投影直接对人脸图像进行检测和眼睛定位是一种常用的算法,但是直接采用该算法会受到背 景、特征等因素的影响,识别准确率较低。提出了一种基于最大类间方差阈值和区域膨胀相结合的检测与定位算法。该算 法首先计算最大类间方差设置阈值,把灰度图像转换为二值图像并检测出人脸区域,然后通过对该人脸区域中的连通区域 进行膨胀及连通性处理,精确定位眼睛坐标。实验表明,此算法可靠,具有较好的识别效果。(face recognized)

包括logistic映射，tent映射，Henon映射，Kent映射的Matlab程序及图像。(Including the logistic map, tent map, Henon mapping Kent Matlab program and image mapping.)