CAJ | 학술논문

随着嵌入式技术不断普及，小型化、操作自由、交互智能成为新一代人机交互技术的发展趋势，基于嵌入式的手势识别就是其中的一项关键技术。主要是成功将OpenCV移植到嵌入式平台，通过V4L2架构实现USB摄像头的视频采集以及LCD显示，利用OpenCV丰富的视觉库实现色彩空间转换，转换到HSV空间，阈值分割得到二值化图像，最后通过滤波处理、轮廓提取、轮廓树匹配的方法实现手势识别，实践证明通过该方法实现嵌入式系统的手势识别可行。
수착감입식기술불단보급，소형화、조작자유、교호지능성위신일대인궤교호기술적발전추세，기우감입식적수세식별취시기중적일항관건기술。주요시성공장OpenCV이식도감입식평태，통과V4L2가구실현USB섭상두적시빈채집이급LCD현시，이용OpenCV봉부적시각고실현색채공간전환，전환도HSV공간，역치분할득도이치화도상，최후통과려파처리、륜곽제취、륜곽수필배적방법실현수세식별，실천증명통과해방법실현감입식계통적수세식별가행。
With the growing popularity of embedded technology, miniaturization, operation freedom and interactive intelligence has become the de-velopment trend of the next generation of human-computer interaction technology, gesture recognition technology based on embedded sys-tem is one of the key technologies. Ports OpenCV to the embedded platform successfully. Through V4L2 architecture enables USB camera video capture and LCD display, uses OpenCV rich visual library achieves color space conversion, the conversion to HSV space, obtains threshold to binary image. Through the filtering process, contour extraction, contour tree matching method to achieve the gesture recogni-tion, experience proves the feasibility of the embedded systems gesture recognition method.