wavelet

自己编写fft2和ifft2程序并进行快速傅里叶变换试验(fft2 and ifft2)

对遥感图像进行无监督分类，形成决策树，实现地物的标定(Unsupervised classification of remote sensing images to form a decision tree, the calibration of surface features)

计算机与机器视觉：理论，算法与实践，第四版；E. R. DAVIES 著；高清文字版(Computer and Machine Vision:Theory, Algorithms,Practicalities Fourth Edition E. R. DAVIES)

说明：  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).)

用于图像融合的IHS变换方法的比较的相关论文(IHS transform for image fusion method of comparison of the relevant papers)

利用Qt编写的简易图像处理程序，包括图像的边缘检测、放大缩小和几个常用的处理功能。(QT image processing procedures, including edge detection, zoom and several common processing functions.)

SIFT算法提取关键点，SIFT算法提取关键点(SIFT algorithm to extract key points, SIFT algorithm to extract key points)

基于Powell和互信息的多分辨率图像配准(Image registration Powell and mutual information)

图像修复中基于一幅图像的改进的Criminisi方法的近，算法的时间大大提高，修改效果也不错。(Time based on the improved image Criminisi near algorithm greatly improve image repair, modify the effect is also good.)

运用Soble,,Roberts算子,Prewitt算子,Log算子,Canny算子j进行边缘检测 (Using Sobel ,, Roberts operator, Prewitt operator, Log operator, Canny edge detection operator )