Based on the linearmemory construction of digital image, this paper presents a method of fast realization of image rotation.
基于图像的线性存储结构,提出了一种图像旋转的快速实现方法。
2
By using CUDA's texture from linearmemory, over-bound accessing is avoided. For reducing access of global memory, the algorithm employs CUDA's Shared memory.
算法通过运用线性存储器纹理解决了越界问题,并借助CUDA的共享内存以减少对全局内存的访问次数。
3
In this way, a large pool of memory can be addressed as if it were linear.