Abstract：In multimedia applications,the image data processing occupies a very important position.With the rapid development of embedded systems,how to accomplish the real-time processing of images in embedded systems has become an extremely difficult challenge.Prewitt edge detection algorithm is a typical image edge detection algorithm,and it is widely used in image processing.In order to improve the computational speed of Prewitt operator in embedded systems,an efficient Prewitt edge detection algorithm which is based on NEON coprocessor is proposed.This algorithm can effectively reduce the amount of redundant computation and image data reading,which improves the computational efficiency.In order to verify the results,a platform which is based on Xilinx Zedboard development board by using Verilog language is built.On this platform,the experimental results show that the speed of embedded systems by using this algorithm is almost the same as the platform which has CPU and GPU.
2018, Vol. 39 
