基于计算机视觉的花生霉变程度检测
|
摘要
花生是重要的油料作物,且富含蛋白质和维生素,具有很高的食用价值。黄曲霉和寄生曲霉能引起花生的霉变,产生强致癌物质黄曲霉素,对人类健康产生严重的威胁。通过对霉变程度的检测,可以为霉变花生的清选提供条件,极大地提高花生食用的安全性。为此,基于计算机视觉技术,用相机拍摄花生的图像,采用维纳滤波处理去除噪音,用B分量进行图像分割后获得目标区域图像。选用H颜色分量作为反映花生霉变程度的特征参数,根据设定的阈值评判霉变等级。该方法对花生霉变程度检测的准确率超过93%,处理单张图像耗时1 s,可以满足实时检测的要求,为花生的在线分选提供了技术支撑。
Peanuts are important oil crops. They are also rich in protein and vitamins and have a high food value. Aspergillus flavus and Aspergillus parasiticus can cause mildew in peanuts,producing a strong carcinogen aflatoxin,which poses a serious threat to human health. Through the detection of moldiness,it can provide conditions for the selection of moldy peanuts and greatly improve the safety of peanuts. This paper is based on computer vision technology,using a camera to take pictures of peanuts,using Wiener filtering to remove noise,and using B-components for image segmentation to obtain images of the target area. The H color component was selected as a characteristic parameter reflecting the degree of mildew in peanut,and the mildew grade was judged according to the set threshold value. The accuracy of the method for the detection of mildew degree of peanut exceeds 93%,and it takes 1 s to process the single-sheet image,which can meet the requirements of real-time detection and provide technical support for on-line sorting of peanuts.
Peanuts are important oil crops. They are also rich in protein and vitamins and have a high food value. Aspergillus flavus and Aspergillus parasiticus can cause mildew in peanuts,producing a strong carcinogen aflatoxin,which poses a serious threat to human health. Through the detection of moldiness,it can provide conditions for the selection of moldy peanuts and greatly improve the safety of peanuts. This paper is based on computer vision technology,using a camera to take pictures of peanuts,using Wiener filtering to remove noise,and using B-components for image segmentation to obtain images of the target area. The H color component was selected as a characteristic parameter reflecting the degree of mildew in peanut,and the mildew grade was judged according to the set threshold value. The accuracy of the method for the detection of mildew degree of peanut exceeds 93%,and it takes 1 s to process the single-sheet image,which can meet the requirements of real-time detection and provide technical support for on-line sorting of peanuts.
引文
[1]时焦,徐宜民.污染花生真菌毒素的研究现状[J].农牧情报研究,1989(12):21-26.
[2]邢君.外贸花生的质量标准与改进措施[J].安徽农学通报,2001,7(1):43-45.
[3]孙春梅,李绍伟,任丽,等.花生品种品质分析及品质育种方向[J].安徽农业科学,2005,33(11):2005-2006.
[4]王叶群,姚刚,张绍英.污染黄曲霉毒素花生的检测及分选技术研究进展[J].农业工程,2014,4(6):59-63.
[5]惠国华,倪彧.基于电子鼻系统的粮食霉变检测方法研究[J].中国食品学报,2011,11(5):162-168.
[6]李雅丽,刘阳.霉变花生光电分选技术应用现状及发展趋势[J].农业机械,2012(12):50-53.
[7]蒋雪松,刘鹏,沈飞,等.衰减全反射-傅里叶变换红外光谱法对花生仁霉变的分析[J].食品科学,2017,38(12):315-320.
[8]乔小军,蒋金豹,李辉,等.高广谱图像识别霉变花生的光谱特征分析与指数模型构建[J].广谱学与广谱分析,2018,38(2):535-538.
[9]何东健,杨青,薛少平,等.果实表面颜色计算机视觉分级技术研究[J].农业工程学报,1998,14(3):202-205.
[10]沈维政,张长利,刘振恒.基于计算机视觉的田间杂草识别方法研究[J].农机化研究,2006,28(7):163-165.
[11]柴阿丽,李宝聚,网倩,等.基于计算机视觉技术的番茄叶片叶绿素含量的测定[J].园艺学报,2009,36(1):45-52.
[12]王彦辉,赵培琨,边东良.基于计算机视觉的瓜果采摘系统的运营研究[J].农机化研究,2018,40(1):200-203.
[13]陈红,熊利荣,胡筱波,等.基于神经网络与图像处理的花生仁霉变识别方法[J].农业工程学报,2007,23(4):158-161.
[14]陈红,吴谋成,熊利荣,等.霉变花生的计算机视觉识别[J].农业机械学报,2008,39(1):110-113.
[15]韩仲志,赵友刚.基于计算机视觉的花生品质分级检测研究[J].中国农业科学,2010,43(18):3882-3891.
[2]邢君.外贸花生的质量标准与改进措施[J].安徽农学通报,2001,7(1):43-45.
[3]孙春梅,李绍伟,任丽,等.花生品种品质分析及品质育种方向[J].安徽农业科学,2005,33(11):2005-2006.
[4]王叶群,姚刚,张绍英.污染黄曲霉毒素花生的检测及分选技术研究进展[J].农业工程,2014,4(6):59-63.
[5]惠国华,倪彧.基于电子鼻系统的粮食霉变检测方法研究[J].中国食品学报,2011,11(5):162-168.
[6]李雅丽,刘阳.霉变花生光电分选技术应用现状及发展趋势[J].农业机械,2012(12):50-53.
[7]蒋雪松,刘鹏,沈飞,等.衰减全反射-傅里叶变换红外光谱法对花生仁霉变的分析[J].食品科学,2017,38(12):315-320.
[8]乔小军,蒋金豹,李辉,等.高广谱图像识别霉变花生的光谱特征分析与指数模型构建[J].广谱学与广谱分析,2018,38(2):535-538.
[9]何东健,杨青,薛少平,等.果实表面颜色计算机视觉分级技术研究[J].农业工程学报,1998,14(3):202-205.
[10]沈维政,张长利,刘振恒.基于计算机视觉的田间杂草识别方法研究[J].农机化研究,2006,28(7):163-165.
[11]柴阿丽,李宝聚,网倩,等.基于计算机视觉技术的番茄叶片叶绿素含量的测定[J].园艺学报,2009,36(1):45-52.
[12]王彦辉,赵培琨,边东良.基于计算机视觉的瓜果采摘系统的运营研究[J].农机化研究,2018,40(1):200-203.
[13]陈红,熊利荣,胡筱波,等.基于神经网络与图像处理的花生仁霉变识别方法[J].农业工程学报,2007,23(4):158-161.
[14]陈红,吴谋成,熊利荣,等.霉变花生的计算机视觉识别[J].农业机械学报,2008,39(1):110-113.
[15]韩仲志,赵友刚.基于计算机视觉的花生品质分级检测研究[J].中国农业科学,2010,43(18):3882-3891.