基于视频的手写数字识别
    点此下载全文
引用本文:王敏,蒋海峰.基于视频的手写数字识别[J].计算技术与自动化,2015,(2):92-94
摘要点击次数: 1783
全文下载次数: 59
作者单位
王敏,蒋海峰 (南京理工大学 自动化院,江苏 南京210094) 
中文摘要:随着计算机技术的不断发展,视频跟踪技术越来越成为计算机领域中研究的热点。视频跟踪技术的研究涉及范围很多,包括视频图像处理、模式识别以及人工智能等,具有较强的研究价值。手势检测识别技术作为一种基于计算机视觉的新型人机交互方式,是其中备受瞩目的研究和应用技术之一。文章采用一种简单高效的颜色直方图对目标(红色手指)进行主色定位,并在图像序列中进行目标区域提取,得到运动轨迹,进行手写数字识别。最后利用八段视频验证了该方法的简单高效,并能成功进行实时跟踪与识别。
中文关键词:颜色直方图  跟踪  手写数字识别
 
Recognition of Handwritten Numeral Based on Video
Abstract:With the continuous development of computer technology, video tracking technology has become a hot research field of computer. Study of video tracking has strong research value, including video and image processing, pattern recognition and artificial intelligence. As a new type of human-computer interaction based on computer vision, gesture detection and identification technology are the high-profile research and applied technology. This article positioned the target (red fingers) with the main color by using a simple and efficient color histogram, extracted the target area in the image sequence to get the motion trajectory, and identified handwritten numeral. Finally, experiment on eight videos shows that this method is simple and efficient. With this method, real-time tracking and gesture recognition are successful.
keywords:color histogram  tracking  handwritten numeral recognition
查看全文   查看/发表评论   下载pdf阅读器