Seed sizing from images of non-singulated grain samples
Seed sizing from images of non-singulated grain samples
复制标题
根据非单一谷物样品的图像确定种子大小
DOI：
--
复制
发表时间：
2005
期刊：
影响因子：
--
通讯作者：
M. Symons
中科院分区：
文献类型：
--
作者：
M. Symons
Shahin，M.A.和西蒙斯，S.J.2005。根据非单粒化颗粒样品的图像确定种子大小。加拿大生物系统工程/Le génie des biosystèmes Au Canada 47：3.49 3.55。图像分析是一个强有力的候选人开发自动测量系统的种子大小。本文提出了一种基于图像分析的非单粒化种子粒径测量方法。IA的方法进行了测试，对四种类型的不同的种子大小，形状和颜色的颗粒。IA测定的球形和非球形种子的平均种子大小和大小分布与手动筛分测定的种子大小密切匹配。IA方法被认为是高度可重复的，比筛分方法快得多。两种方法的操作员之间和操作员内部差异无统计学意义（p > 0.05）。IA方法测量一个样品需要不到30秒，而手动筛分需要超过10分钟。
Shahin, M.A. and Symons, S.J. 2005. Seed sizing from images of non-singulated grain samples. Canadian Biosystems Engineering/Le génie des biosystèmes au Canada 47: 3.49 3.55. Image analysis is a strong candidate for developing automated measurement systems for seed sizing. A unique image analysis (IA) based approach to measuring seed size from non-singulated samples is presented in this paper. The IA method was tested on four types of grains differing in seed size, shape, and colour. IA determined seed sizes closely matched those determined by manual sieving both in average seed size and the size distribution for seeds of spherical as well as non-spherical shape. The IA method was found to be highly repeatable and much faster than the sieving method. Between and within operators’ differences were statistically insignificant (p > 0.05) for both the methods. The IA method took less than 30 seconds to measure one sample as opposed to over 10 minutes for manual sieving.