Development of a computer vision system and novel evaluation criteria to characterize color and appearance of rice

Journal article

Authors/Editors

Strategic Research Themes

No matching items found.

Publication Details

Author list: Jinorose M., Prachayawarakorn S., Soponronnarit S.

Publisher: Taylor and Francis Group

Publication year: 2010

Journal: Drying Technology (0737-3937)

Volume number: 28

Issue number: 9

Start page: 1118

End page: 1124

Number of pages: 7

ISSN: 0737-3937

eISSN: 1532-2300

URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-77956632039&doi=10.1080%2f07373937.2010.506174&partnerID=40&md5=cdbadfff7af2d331f073308c863b6faa

Languages: English-Great Britain (EN-GB)

View in Web of Science | View on publisher site | View citing articles in Web of Science

Abstract

Rice quality evaluation, both during and after drying, is traditionally performed by visual inspection. However, this is a tedious, time-consuming, and subjective task. More importantly, there is hardly any standard objective method to effectively evaluate the quality of rice. Proper inspection methods, which would allow a better control of the rice drying process, are thus in great demand. Image analysis has recently emerged as one of the most promising techniques for quality analysis of various food products, including rice. However, in order to successfully apply image analysis to rice quality evaluation, an appropriate means to characterize a rice kernel is first needed. This study aimed to develop effective but simple computer-vision algorithms along with novel evaluation criteria that could be used to simultaneously inspect various visual qualities of rice, including grain contour, size, and color in terms of various types of kernel damage, viz. undermilled, red, yellow, and chalky kernels. It was found that the developed algorithms could be used to assess some dimensional parameters such as the major axis, minor axis, and projected area of rice kernels effectively. Through the use of the multivariate discriminant analysis, it was found that the hue, saturation, and value (HSV) color space could be used to evaluate various kernel defects, including kernel discolorations and chalkiness, that are otherwise difficult to assess satisfactorily. ฉ 2010 Taylor & Francis Group, LLC.

Keywords

Color systems, Kernel discoloration, Multivariate discriminant analysis, Quality evaluation