Linux下基于CUDA的RANSAC实现，可以转到windows下的VS2010下运行，需要OpenCV支持。可以支持64bit编译环境。具有很好的参考价值，需要花时间调试。-Based on Linux with CUDA RANSAC, you can go to run windows under VS2010, you need to OpenCV support. You can support 64bit build environment. Have a good referen

使用NVIDIA的CUDA并行架构下的模板匹配，效率比openCV的有很大提升-Template matching using the NVIDIA CUDA parallel architecture, enhance efficiency than openCV

cuda并行计算去雾，在linux环境下，使用opencv库，nvcc hazy.cu 运行-cuda parallel computing to fog, in the linux environment, using opencv library, nvcc hazy.cu run

利用cuda进行的简单的加法操作，需要opencv的支持-Cuda conducted utilizing simple addition, you need to support the opencv