引用本文:[点击复制]
[点击复制]
【打印本页】【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

←前一篇|后一篇→

过刊浏览    高级检索

本文已被:浏览次   下载 本文二维码信息
码上扫一扫!
基于镜头检测的成熟期水稻图像处理算法研究
王轲,邵陆寿
0
(安徽农业大学经济技术学院,合肥 230036; 安徽农业大学工学院,合肥 230036)
摘要:
快速处理视频信息,实时获取水稻生长密度,是实现水稻联合收割机喂入量实时控制的关键。采用模板算法,引入基于镜头检测的视频挖掘技术,提取视频信息中的关键帧作为水稻密度突变检测依据。结果表明,镜头检测技术的平均每帧检测时间是0.05 s,比传统静态图像算法提取作物的密度特征速度快16倍;模板算法可有效消除田间作物品种、生长态势、光照等自然因素的影响,算法具有通用性。
关键词:  水稻密度  镜头检测  喂入量  联合收割机
DOI:CNKI:34-1162/S.20111025.1029.020
基金项目:国家科技支撑计划项目(2009BADA6B02)资助。
An algorithm study of image processing for mature rice based on shot detection
WANG Ke,SHAO Lu-shou
(School of Economics and Technology, Anhui Agricultural University, Hefei 230036; School of Engineering, Anhui Agricultural University, Hefei 230036)
Abstract:
In order to realize real-time controlling of feed quantity, fast access is the key to get large quantities of image video information and rice growth density in real time. This paper adopted template algorithm and video mining technology based on shot detection. We extracted the key frames from video information as the basis of rice in density detection. The results indicated that real-time detection of rice crop growth density by using shot detection was realizable and the average detection time is 0.05s. The running time of the template algorithm was 16 times faster than that of the traditional algorithms of image. The template algorithm reduce or eliminate effect of natural factors such as the crop strains, growth situation and illumination,so the algorithm could be universality.
Key words:  density of rice  shot detection  feeding volume  combine harvester

用微信扫一扫

用微信扫一扫