基于图像处理的小麦穗长和小穗数同步测量  被引量:14

Synchronous measurement of wheat ear length and spikelets number based on image processing

在线阅读下载全文

作  者:路文超 罗斌[1,2] 潘大宇 赵勇[1,2,4] 于春花 王成[1,2] 

机构地区:[1]北京农业智能装备技术研究中心,北京市100097 [2]农业部农业信息技术重点实验室,北京市100097 [3]国家农业智能装备工程技术研究中心,北京市100097 [4]农业智能装备技术北京市重点实验室,北京市100097

出  处:《中国农机化学报》2016年第6期210-215,共6页Journal of Chinese Agricultural Mechanization

基  金:863课题(2012AA10A503);公益性行业(农业)科研专项(201203026);北京市自然科学基金(4142019)

摘  要:小麦穗长和小穗数能够直接反映小麦产量,是育种家非常关心的两个穗部形态参数。为快速测量这两个参数,本文提出一种基于图像处理技术的穗长和小穗数同步测量方法,首先采用形态学处理算法去除麦芒得到麦穗主部图像,然后采用拟合麦穗主部中心曲线的方法来计算麦穗长度,最后采用将曲线穿过小穗区域并统计灰度差异的方法计算小穗数。针对4个品种的50个麦穗进行测量试验,以验证算法性能。试验结果表明,本文穗长和小穗数同步提取算法不仅适用于较直立的麦穗,而且适用于弯曲的麦穗;和人工用直尺沿穗身测量穗长的方法相比,本文穗长提取算法测量结果的平均绝对误差为0.47cm,相对误差为3.95%;和人眼计数测量小穗数的方法相比,本文小穗数提取算法测量结果的平均绝对误差为1.6(≈2),相对误差为7.89%,能有效同步测量穗长和小穗数。Wheat ear length and spikelets number directly reflect the yield of wheat,and the two parameters are greatly concerned by wheat species breeding and test experts.This paper proposed a synchronous measurement method of wheat ear length and spikelets number based on image processing,firstly,the main part of ear was extracted through the morphological algorithm,secondly,the ear length was calculated by detecting the center curve of the ear main part,at last,spikelets number was further computed by the curve passing through the highlighted spikelets' regions.In this study,50 random wheat ears of 4wheat varieties were tested,the test results demonstrated that our method not only applied to the upright ears but also applied to the curved ears;compared with the actual values measured by rulers,the mean absolute error of our ear length measure method was 0.47 cm,the average relative error was 3.95%;compared with the actual values measured by visual inspection,the mean absolute error of our spikelets number measure method was 1.6(≈2),the average relative error was 7.89%;the above showed that the wheat ear length and spikelets number could be effectively measured by our method.

关 键 词:图像处理 穗长 小穗数 同步测量 形态学处理 曲线拟合 

分 类 号:S24[农业科学—农业电气化与自动化]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象