期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
基于光饱和影响校正的作物叶绿素分布光谱成像检测 被引量:1
1
作者 孙红 邢子正 +4 位作者 乔浪 龙耀威 高德华 李民赞 Qin Zhang 《光谱学与光谱分析》 SCIE EI CAS CSCD 北大核心 2019年第12期3897-3903,共7页
叶绿素含量是作物光合能力与营养评价的重要指标,因此快速检测作物叶绿素含量与分布可为作物营养动态分析与长势评估提供支持。基于RGB(Red,Green,Blue)和NIR(Near Infrared)多光谱图像的获取,开展玉米作物营养状态分布光谱学成像检测... 叶绿素含量是作物光合能力与营养评价的重要指标,因此快速检测作物叶绿素含量与分布可为作物营养动态分析与长势评估提供支持。基于RGB(Red,Green,Blue)和NIR(Near Infrared)多光谱图像的获取,开展玉米作物营养状态分布光谱学成像检测。构建了多光谱图像采集平台获取RGB和NIR图像,研究了基于光饱和校正算法的RGB图像的光饱和校正与NIR图像去噪方法,通过图像的匹配分割,冠层的提取校正,建立了基于冠层图像的作物SPAD值检测模型与分布成图。采集15株玉米植株RGB-NIR图像,并同步获取不同植株,不同位置共68个叶绿素含量指标SPAD值。首先对RGB图像进行光饱和校正,再对NIR图像进行滤波和图像增强,其次对RGB和NIR图像进行了SURF(speeded-up robust features)和RANSAC(random sample consensus)图像匹配,利用RGB图像的颜色特征,采用ExG(Extra Green)和OTSU算法生成分割掩模,对RGB图像和NIR图像进行分割提取,提取图像的R,G,B和NIR分量,利用4阶灰度板进行反射率校正,然后计算作物图像中像素级P SPAD值,并建立图像P SPAD值与叶绿素仪SPAD值的拟合模型,最后绘制作物SPAD分布图。通过HSI(Hue,Saturation,Intensity)彩色模型中的I分量直方图对比去饱和前后光分布范围,以作物SPAD值分布图验证光饱和校正算法对作物叶绿素含量分布检测提升的效果。RGB图像光饱和校正前I分量集中在[140~180]之间,光饱和校正后的RGB图像I分量集中在[85~130]之间,校正了相机成像时产生模糊和RGB图像饱和。对分割后的RGB图像和NIR图像提取R,G,B,NIR分量进行4阶灰度板校正,相关系数分别为0.829,0.828,0.745和0.994,进而生成R,G,B和NIR四波段的反射率伪彩色图像,反射率R NIR>R G>R R>R B。体现了作物的在蓝光和红光区域吸收光,在绿光区域和近红外区域反射光的光谱特性。校正前后的R和NIR分量反射率计算图像P SPAD值拟合叶绿素含量指标SPAD值的模型结果显示,校正前R 2为0.3326,校正后R 2为0.6193,绘制作物的SPAD特征分布图,可为作物的营养动态快速分析与分布检测提供技术支持。 展开更多
关键词 光饱和校正 多光谱分析 作物检测 叶绿素分布 图像处理
下载PDF
Detection system of smart sprayers: Status, challenges, and perspectives 被引量:6
2
作者 Sun Hong Li Minzan Qin Zhang 《International Journal of Agricultural and Biological Engineering》 SCIE EI CAS 2012年第3期10-23,共14页
A smart sprayer comprises a detection system and a chemical spraying system.In this study,the development status and challenges of the detection systems of smart sprayers are discussed along with perspectives on these... A smart sprayer comprises a detection system and a chemical spraying system.In this study,the development status and challenges of the detection systems of smart sprayers are discussed along with perspectives on these technologies.The detection system of a smart sprayer is used to collect information on target areas and make spraying decisions.The spraying system controls sprayer operation.Various sensing technologies,such as machine vision,spectral analysis,and remote sensing,are used in target detection.In image processing,morphological features are employed to segment characteristics such as shape,structure,color,and pattern.In spectral analysis,the characteristics of reflectance and multispectral images are applied in crop detection.For the remote sensing application,vegetation indices and hyperspectral images are used to provide information on crop management.Other sensors,such as thermography,ultrasonic,laser,and X-ray sensors,are also used in the detection system and mentioned in the review.On the basis of this review,challenges and perspectives are suggested.The findings of this study may aid the understanding of smart sprayer systems and provide feasible methods for improving efficiency in chemical applications. 展开更多
关键词 smart sprayer target detection weed control disease detection chemical application
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部