欧亚国际

较近，来自Aarhus大学的Birte 教授研究团队发表了题为Classification of Processing Damage in Sugar Beet (Beta vulgaris) Seeds by Multispectral  Image Analysis 的文章，对多光谱成像技术在种子质量控制的应用进行了深入研究。VideometerLab 多光谱成像系统是的光谱、计算机等技术集成设备，体现了近视距多光谱研究的较高水准，广泛为先进机构如ISTA等等广泛使用。

Classification of Processing Damage in Sugar Beet (Beta vulgaris) Seeds by Multispectral  Image Analysis

Zahra Salimi and Birte Boelt *

Department of Agroecology, Aarhus University, 4200 Slagelse, Denmark; z.salimi@agro.au.dk

*  Correspondence: bb@agro.au.dk

Received: 17 April 2019; Accepted: 16 May 2019; Published: 22 May 2019

Abstract: The pericarp of monogerm sugar beet seed  is rubbed off during  processing  in  order to produce uniformly  sized seeds ready  for  pelleting.  This process  can lead to mechanical  damage, which  may cause  quality  deterioration of  the  processed seeds. Identification of  the mechanical damage and classification of the severity of the injury is important and currently time consuming, as visual inspections by trained analysts are used. This study aimed to find alternative seed quality assessment methods by evalsuating a machine vision technique for the classification of five damage types in  monogerm sugar  beet seeds. Multispectral  imaging (MSI)  was employed using the VideometerLab3 instrument and  instrument software. Statistical analysis of  MSI-derived data produced a model, which had an average of 82% accuracy in classification of 200 seeds in the five damage classes. The first class contained seeds with the potential to produce good seedlings and the model was designed to put more limitations on seeds to be classified in this group. The classification accuracy of  class one  to five was  59,  100,  77,  77 and 89%, respectively. Based  on  the results we conclude that MSI-based classification of mechanical damage in sugar beet seeds is a potential tool for future seed quality assessment.

Keywords: machine vision; mechanical damage; prediction model; seed quality; seed polishing