Aimed at ARM-based embedded image processing

the ARM Cortex-A series of processors are integrated the NEON coprocessor engine which is based on the SIMD instruction architecture

enabling an instruction to parallelize multiple data simultaneously

which can accelerate multimedia and signaling processing algorithms (such as video codecs

image processing techniques) effectively. In this paper

we design a video processing system based on FPGA

which can make full use of NEON technology to optimize the Sobel algorithm

compared with the way of ordinary software method to implement the algorithm

the results show that NEON technology accelerates image processing significantly

decoded the video more fluent.