摘要
为实现对籽棉品质进行快速、准确的检测,我们设计了一种籽棉颜色级检测仪,通过检测籽棉的反射率和黄度,判断出籽棉的颜色级,为籽棉收购、结价及加工过程智能化控制提供技术支持.由于传感器及电路系统存在一定的误差,开发的籽棉颜色级检测仪,在实际使用前要进行适当校正.本文在软件校正等方面进行了针对性研究与设计.经过人工神经网络校正后,检测数据的准确性显著提高且离散性显著降低.
In order to detect the quality of seed cotton quickly and accurately,we designed a Color-Level detector for seed cotton.Through the detection of seed cotton color(Rd,+b),the color-level of seed cotton can be determined,which provided technical support for seed cotton trading,setting price and intelligent control of processing.Due to the errors of sensor and circuit system,the designed seed cotton color-level detector should be corrected properly before practical use.This paper focuses on software correction by artificial neural network.After the correction through the artificial neural network,the accuracy of the detection result is significantly improved and the discreteness is significantly reduced.
作者
梁后军
谢睿
LIANG Hou-jun;XIE Rui(College of Management Science and Engineering,Anhui University of Finance and Economics,Bengbu 233000,Anhui,China)
出处
《山西师范大学学报（自然科学版）》
2020年第1期50-56,共7页
Journal of Shanxi Normal University(Natural Science Edition)
基金
安徽省自然科学基金(1808085MC88).
关键词
籽棉品质检验
颜色级
颜色传感器
人工神经网络
quality test of seed cotton
color level
color sensor
artificial neural network
