Saliency_Detection

说明：  可见光影像的密集匹配，很经典的算法，效果也非常理想(image matchingimage matching)

盲信号分离的fastICA算法，读取具有混合干扰的4幅图像，处理后显示分离后图像(Blind Source Separation Algorithm fastICA, read with mixed disturbance of four images, processed images show that after the separation)

利用GPU加速的SIFT算法，实现影像匹配，并利用RANSAC进行粗差的剔除(GPU-accelerated use SIFT algorithm to achieve image matching, and use RANSAC removing coarse poor)

LVQ(学习矢量量化)算法：它是Kohonen的有监督学习的扩展形式,融合了自组织和有导师监督的技术,学习方法是竞争的,但(LVQ (Learning Vector quantization) algorithm: it is an extended form of supervised learning of Kohonen, which combines self-organizing and supervised techniques. The learning method is competitive, but the)

说明：  多尺度多方向性的剪切波Shearlet变换优于小波变换，里面包含很多例子，可以直接运行，可以用于图像去噪，分割，增强等，是一个优于小波的图像稀疏表示工具(Shearlet transform of multi-scale and multi-directional shear wave is better than wavelet transform. It contains many examples and can run directly. It can be used for image denoising, segmentation and enhancement. It is an image sparse representation tool better than wavelet transform.)

说明：  视频追踪的框架算法，实现了基于meanshift和feature based的跟踪算法(Framework for video tracking algorithm meanshift based and feature based tracking algorithm)

图像移动，随着鼠标移动图像，可以自由在空间内平移(image auto move)

说明：  采用生成对抗网络（GAN）实现图像去模糊，用于图像的清晰化处理(The generated countermeasure network (GAN) is used to remove the blur of the image, which is used to clear the image)

计算图像的均值和方差函数，matlab编写的，可以添加到自己的图像处理中(Calculated the mean and variance of the image function, matlab prepared, can be added to their own image processing)

说明：  利用brisk算法对两张图片进行特征检测，并且实现图像拼接。(The brisk algorithm is used to detect the features of the two images and realize the image mosaic.)