caffe-ssd.tar

RAT power toy (KOL)

高斯分布二维的数据的产生和画图函数，供数据仿真分析用(Two-dimensional Gaussian distribution of data generation and drawing functions, for simulation data analysis)

模糊图像去模糊处理，Amit等人发表的模糊核的估计算法，图像去模糊，得到清晰图像(image deblurring by Amit)

图像超分辨率是指从一序列低分辨率观测图像中重建高分辨率图像。图像超分辨率充分利用了各帧图像 间相似而又不同的信息。(Compared with the model derived from Kolmogorov spectrum, the new model introduces more complete transcendent constraints and is more similar to the physical characteristics of atmospheric turbulence.)

在桌面上显示图像，C++Builder精选编程学习源码，很好的学习资料。(Image display on the desktop, C++Builder select learning programming source code, very good learning materials.)

2009年CVPR最佳论文,何恺明的基于暗通道的图像去雾的相关论文，ppt以及matlab代码. Dark channel prior 是一种对无雾的图像的统计规律：在无雾图像中，大部分图像块中至少有一个颜色通道具有很小的颜色强度值。利用dark channel prior，可以直接从一幅有雾的图像中估计出雾的浓度，并恢复出高质量的去雾图像。(CVPR best paper in 2009 (He Kai Ming Single Image Removal Haze is the Using the Dark Channel) paper, ppt and the matlab code, it is)

快速TV算法用于图像去模糊、图像去噪等，其中TV模式包括两种(TV went fast algorithm for image blur, image denoising, which includes two TV mode)

关于计算机视觉和图像的matlab工具箱。包括相位一致性特征，空间特征检测，积分图像，非极大值抑制和滞后阈值，边链接和线段拟合，去噪，等(MATLAB and Octave Functions for Computer Vision and Image Processing。such as • Feature Detection via Phase Congruency • Spatial Feature Detection • Integral Images • Non-Maxima Suppression and Hysteresis Thresholding • Edge Linking and Line Segment Fitting • Test Grating for Edge Detection • Image Denoising • Surface Normals to Surfaces • Scalogram Calculation • Anisotropic diffusion • Grey Scale Transformation and Enhancement • Frequency Domain Transformations • Functions Supporting Projective Geometry • Feature Matching • Model Fitting and Robust Estimation • Fingerprint Enhancement • Interesting Synthetic Images • ASCII Image Generation • Rotation Transforms • Image Blending • Colourmaps and colour conversions • Image Display, Image Writing and Miscellaneous )

IMU姿态跟踪，利用粒子滤波器实现，可用于机器人姿态确定(IMU orientation tracking algorithm using particle filter.)

说明：  将基于小波变换的嵌入水印后的宿主图像进行压缩处理再提取灰度水印。(will be based on the wavelet transform the embedded watermark image of the host compressed again from Gray watermark.)