智能驾驶汽车机器视觉关键技术分析

Analysis of Key Technology of Machine Vision for Intelligent Driving Vehicle

文章对智能驾驶汽车机器视觉关键技术中的图像采集、图像处理和图像分析三个部分进行研究。在图像采集技术中,重点研究带电耦合器件(CCD)图像传感器和互补金属氧化物半导体(CMOS)图像传感器的特点及其工作原理,并分别介绍基于现场可编程门阵列(FPGA)和数字信号处理机(DSP)的图像采集卡的工作特点及适用环境。在图像处理技术中,对感兴趣区域提取、图像灰度化处理、图像滤波去噪和边缘检测等环节进行研究。在图像分析技术中,重点对比传统的数字图像识别技术和基于深度学习的图像识别技术。结合对三个部分工作原理与特点的研究,对智能驾驶汽车机器视觉技术未来的发展趋势进行展望,提出人工智能化、车联网、车辆通信系统和多传感器化等发展构想。

The key technologies of machine vision for intelligent driving vehicle are studied in this paper.Machine vision technology is composed of three parts:image acquisition,image processing and image analysis.Firstly,it introduces the whole procedure of the image acquisition,focuses on the analysis of the characteristics of charge coupled device(CCD)image sensor and complementary metal oxide semiconductor(CMOS)image sensor and their working principles.Furthermore,the application of image acquisition card based on field programmable gate array(FPGA)and digital singnal processor(DSP)has been investigated.Secondly,the key technologies of the image processing,including the region of interest extraction,image gray processing,image filtering denoising and edge detection are studied.Thirdly,in terms of the image analysis technology,the traditional digital image recognition technology and deep learning-based image recognition technology are compared.Combined with the research results of this paper,the future development trend of intelligent driving vehicle machine vision technology is prospected,and the development ideas of artificial intelligence,vehicle networking,vehicle communication system and multi-sensor are expounded.