面向人形机器人的FPGA综合图像处理系统

谢天舒; 刘远光; 徐尚睿; 李泽林; 黄永嘉; 张弘; 娄永乐

doi:10.20193/j.ices2097-4191.2025.0094

PDF(10252 KB)

集成电路与嵌入式系统 ›› 2026, Vol. 26 ›› Issue (2) : 71-80. DOI: 10.20193/j.ices2097-4191.2025.0094

第九届全国大学生集成电路创新创业大赛优秀作品专刊

面向人形机器人的FPGA综合图像处理系统

作者信息 +

Design of image processing system for humanoid robots based on FPGA

Author information +

文章历史 +

摘要

为解决ARM架构延迟高和FPGA方案功能单一的问题,设计了一套基于FPGA与PC协同架构的图像处理系统。系统集成对亮度、对比度和色温的调节,绿幕抠图,肤色ROI,信号灯ROI提取和无效区域剔除等功能,上位机通过Python Flask框架构建Web界面,实现参数配置与结果展示,并扩展了手势识别功能。通过USB-UART链路实现数据交互,核心模块处理速度稳定在560 Mb/s,大幅提升了图像处理效率,满足实时性需求。该系统为人形机器人视觉前端提供高质量图像输入,适应低光和遮挡场景,具有重要的应用价值。

Abstract

To address the high latency of ARM architecture and the limited functionality of FPGA solutions, a collaborative architecture-based image processing system combining FPGA and PC is designed. The system integrates functions such as brightness, contrast, and color temperature adjustment, green screen matting, skin tone ROI, traffic light ROI extraction, and invalid region removal. The host computer builds a web interface using the Python Flask framework to implement parameter configuration and result display, and also extends the gesture recognition functionality. Data interaction is achieved through a USB-UART link, and the core module's processing speed remains stable at 560 Mb/s, significantly improving image processing efficiency and meeting real-time requirements. This system provides high-quality image input for humanoid robot vision front-end, adapting to low-light and occlusion scenarios, with broad application value.

导出引用

谢天舒, 刘远光, 徐尚睿, 等. 面向人形机器人的FPGA综合图像处理系统[J]. 集成电路与嵌入式系统. 2026, 26(2): 71-80 https://doi.org/10.20193/j.ices2097-4191.2025.0094

XIE Tianshu, LIU Yuanguang, XU Shangrui, et al. Design of image processing system for humanoid robots based on FPGA[J]. Integrated Circuits and Embedded Systems. 2026, 26(2): 71-80 https://doi.org/10.20193/j.ices2097-4191.2025.0094

中图分类号： TP274 (数据处理、数据处理系统)

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]	Intel Corporation. High-Performance Image Signal Processing and Camera Sensor Pipeline Design on FPGAs[R]. Santa Clara,CA,USA, 2024. 本文引用 [2]

[2]	中国信息通信研究院泰尔系统实验室. 人形机器人产业发展研究报告(2024年)[R], 2024. Telecom System Laboratory, China Academy of Information and Communications Technology (CAICT). Research Report on the Development of Humanoid Robot Industry (2024)[R], 2024(in Chinese). 本文引用 [1]

[3]	BUNDSCHUH S, KUNZE J, KUHNERT K D. Implementation of an FPGA-Based System to Process Images and Match Keypoints on High-Resolution Pictures[J]. Electronics, 2024,13:4774. 本文引用 [3]

[4]

李佩斌. 基于高速串行总线的DSP+FPGA架构图像处理系统设计[J]. 集成电路与嵌入式系统, 2024, 24(12):45-51.

https://doi.org/10.20193/j.ices2097-4191.2024.0033

摘要

采用高性能DSP+FPGA架构可满足嵌入式图像处理系统对大数据量、复杂算法的实时处理需求,传统的DSP+FPGA架构使用并行外部存储器接口作为数据传输接口,走线条数较多,布线难度大,故障点多。采用高速串行总线可解决以上问题,本文提出一种基于高速串行总线的DSP+FPGA架构图像处理系统,采用PCIe总线作为DSP与FPGA之间的图像数据传输通道,SRIO总线作为DSP与DSP之间的数据传输通道,SGMII总线作为DSP与PHY芯片的数据传输通道。高速串行总线使得数据传输率更高,布线更容易,减小了电磁干扰,提高了抗干扰能力。本文设计的系统已在实际场所中部署并稳定运行,验证了设计的可行性和系统的可靠性。

P B

. Design of an Image Processing System Based on DSP+FPGA Architecture with High-Speed Serial Bus[J]. Integrated Circuits and Embedded Systems, 2024, 24(12):45-51(in Chinese).

本文引用 [2]

[5]	ABM REZBAUL ISLAM, ALI ALAMMARI, BILL BUCKLES. Human skin detection:An unsupervised machine learning way[J]. Journal of Visual Communication and Image Representation, 2024,98. 本文引用 [1]

[6]	周先春, 郭亮可, 邹宇, 等. 基于FPGA实现二维卷积滤波器的图像处理[J]. 现代电子技术, 2021, 44(4):17-22. ZHOU X C, GUO L K, ZOU Y, et al. Image Processing of Two-Dimensional Convolution Filter Realized Based on FPGA[J]. Modern Electronic Technology, 2021, 44(4):17-22(in Chinese). 本文引用 [2]

[7]	张强, 孙静, 王威廉, 等. 硬件加速CNN实时图像处理方法[J]. 计算机工程与设计, 2020, 41(6):1581-1585. ZHANG Q, SUN J, WANG W L, et al. Hardware Accelerated CNN Real-Time Image Processing Method[J]. Computer Engineering and Design, 2020, 41(6):1581-1585(in Chinese). 本文引用 [1]

[8]	CHEN L, WANG F, DENG H, et al. A Survey on Hand Gesture Recognition[C]// 2013 International Conference on Computer Sciences and Applications,Wuhan, China, 2013:313-316. 本文引用 [1]

[9]	C LEE, XU Y S. Interactive learning of gestures for human/robot interfaces[C]// Proceedings of IEEE International Conference on Robotics and Automation, Minneapolis,MN, USA, 1996:2982-2987. 本文引用 [1]

[10]	马景. 基于多传感器和Zynq的手势识别系统[D]. 成都: 西南交通大学, 2019. MA J. Gesture Recognition System Based on Multi-Sensors and Zynq[D]. Chengdu: Southwest Jiaotong University, 2019(in Chinese). 本文引用 [2]

[11]	MATIWADE, SUNITA V, DR M R DIXIT. Electronic Support System to Interpret Sign Language of Communication Used by Deaf and Dumb Community[J]. JETIR, 2015(9). 本文引用 [1]

[12]

刘湘, 魏鑫, 陈禾, 等. 应用于ARM-NEON的图像二维卷积高效实现方法[C]// 中国高科技产业化研究会智能信息处理产业化分会. 第十二届全国信号和智能信息处理与应用学术会议论文集, 2018:498-503.

LIU

, WEI

, CHEN

, et al. Efficient Implementation of 2D Image Convolution for ARM-NEON[C]// Proceedings of the 12th National Conference on Signal and Intelligent Information Processing and Applications. Intelligent Information Processing Industrialization Branch, China High-Tech Industrialization Research Association; School of Information and Electronics, 2018:498-503(in Chinese).

本文引用 [1]

[13]	李钢, 李静. 基于FPGA图像处理技术的人脸检测[J]. 工业控制计算机, 2020, 33(3):84-86. LI G, LI J. FPGA-Based Image Processing Technology for Face Detection[J]. Industrial Control Computer, 2020, 33(3):84-86(in Chinese). 本文引用 [1]