A New Automatic Real-Time Crop Row Recognition Based on SoC-FPGA
With the development of articial intelligence technology, agricultural robot plays a signi- cantly important role for agricultural intelligence. Crop row line detection is a critical and fundamental step for agricultural robot navigation. Although there are some crop row lines detection methods, few of them can meet the real-time requirement for agricultural robot under complex elds conditions. In view of this, a real-time crop detection system implemented on a SoC FPGA (System-on-a-Chip Field Programmable Gate Array) is rst proposed in this paper, which contains crop segmentation and crop row detection, where we design parallel pipeline architecture to enhance real-time performance by using line buffer and sliding windows technologies. At the same time, the xed point representation is used to reduce the memory resource in this system. The proposed system is evaluated and implemented on Xilinx Zynq UltraScaleC MPSoC ZCU102 SoC-FPGA. The experimental results show that the proposed system can process an image with 19201080 resolution only within 210 ms with the average accuracy of 89.7%, which satises the real-time requirements of the crop rows recognition.
Crop row recognition, real-time image processing, SoC FPGA