柑橘真菌感染部位的高光谱成像快速检测
发布时间：2019-04-01 15:18
【摘要】：真菌感染是柑橘的一种常见病害,是柑橘腐烂的主要因素,自动化检测出柑橘真菌感染可以有效提高柑橘的商品价值和市场竞争力。运用高光谱成像技术对真菌感染柑橘腐烂部位的缺陷特征进行了快速识别检测。基于ROI提取柑橘真菌感染光谱曲线,对光谱矩阵进行主成分分析,分析权重曲线后得到4个特征波段,分别为615,680,710和725nm,然后对这4波段组合分别做主成分分析,通过分析权重曲线提取到615和680nm两个特征波段,基于这两个特征波段做主成分分析,以第2主成分图像为基础识别柑橘真菌感染部位,识别率达到了100%。高光谱成像技术可用于快速检测柑橘真菌感染引起的腐烂缺陷,为开发水果分级和缺陷检测等相关仪器设备的研究提供了理论方法和依据。
[Abstract]:Fungal infection is a common disease of citrus, it is the main factor of citrus rot, and the automatic detection of citrus fungal infection can effectively improve the commodity value and market competitiveness of the citrus. In this paper, the characteristics of the defect of the citrus rot in the fungal infection were identified by the high-spectral imaging technique. extracting the spectrum curve of the citrus fungus infection based on the ROI, performing principal component analysis on the spectrum matrix, analyzing the weight curve to obtain four characteristic bands,615,680,710 and 725 nm, respectively, and then performing main component analysis on the four-band combination, By analyzing the weight curve to the two characteristic bands of 615 and 680 nm, the main component analysis is made based on the two characteristic bands, and the site of the citrus fungus infection is identified on the basis of the second main component image, and the recognition rate is 100%. The high-spectrum imaging technology can be used to quickly detect the decay defects caused by the infection of the citrus fungi, and provides the theoretical method and the basis for the research of the related instruments and equipment such as the development of the fruit classification and the defect detection.
【作者单位】： 浙江大学生物系统工程与食品科学学院;华东交通大学;
【基金】：国家重大仪器设备开发专项(2014YQ470377) 国家支撑技术项目(2015BAD19B03) 国家自然科学基金项目(61071220) 江西省科技支持项目(20123BDH80014)资助
【分类号】：S436.66;TP391.41
本文编号：2451663
[Abstract]:Fungal infection is a common disease of citrus, it is the main factor of citrus rot, and the automatic detection of citrus fungal infection can effectively improve the commodity value and market competitiveness of the citrus. In this paper, the characteristics of the defect of the citrus rot in the fungal infection were identified by the high-spectral imaging technique. extracting the spectrum curve of the citrus fungus infection based on the ROI, performing principal component analysis on the spectrum matrix, analyzing the weight curve to obtain four characteristic bands,615,680,710 and 725 nm, respectively, and then performing main component analysis on the four-band combination, By analyzing the weight curve to the two characteristic bands of 615 and 680 nm, the main component analysis is made based on the two characteristic bands, and the site of the citrus fungus infection is identified on the basis of the second main component image, and the recognition rate is 100%. The high-spectrum imaging technology can be used to quickly detect the decay defects caused by the infection of the citrus fungi, and provides the theoretical method and the basis for the research of the related instruments and equipment such as the development of the fruit classification and the defect detection.
【作者单位】： 浙江大学生物系统工程与食品科学学院;华东交通大学;
【基金】：国家重大仪器设备开发专项(2014YQ470377) 国家支撑技术项目(2015BAD19B03) 国家自然科学基金项目(61071220) 江西省科技支持项目(20123BDH80014)资助
【分类号】：S436.66;TP391.41
【相似文献】
相关期刊论文 前4条
1 田有文;程怡;吴琼;牟鑫;;农产品病虫害高光谱成像无损检测的研究进展[J];激光与红外;2013年12期
2 王斌;薛建新;张淑娟;;基于高光谱成像技术的腐烂、病害梨枣检测[J];农业机械学报;2013年S1期
3 冯雷;张德荣;陈双双;冯斌;谢传奇;陈佑源;何勇;;基于高光谱成像技术的茄子叶片灰霉病早期检测[J];浙江大学学报(农业与生命科学版);2012年03期
4 郑志雄;齐龙;马旭;朱小源;汪文娟;;基于高光谱成像技术的水稻叶瘟病病害程度分级方法[J];农业工程学报;2013年19期
相关硕士学位论文 前3条
1 邢晓祺;基于高光谱成像技术的玉米螟检测方法研究[D];沈阳农业大学;2016年
2 罗阳;基于NIR高光谱成像技术的长枣虫害及可溶性固形物无损检测研究[D];宁夏大学;2013年
3 陈纳;基于高光谱成像技术油菜菌核病的快速诊断研究[D];浙江大学;2015年
，本文编号：2451663
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2451663.html
