基于计算机视觉的番茄营养元素亏缺识别研究
|
摘要
针对番茄种植中营养元素的亏缺,肉眼不易进行识别判断的问题,以番茄亏缺氮、镁营养元素为研究对象,利用CDD摄像机采集研究图像,将图像进行处理后,提取分割出可以表现亏缺氮、镁的特征图像,提取颜色特征和纹理特征,并通过遗传算法进行优化。同时,将优化的特征进行组合分析,以此建立特征模型,并确定特征向量用于分析提取出来的特征参数,建立的特征模型,并采用二叉树形式对番茄缺素识别进行研究。仿真试验结果表明:番茄种植中,采用计算机视觉技术识别亏缺氮、镁营养元素,识别准确率可以满足生产需要。种植户可以根据检测结果对番茄进行区别施肥,既能满足番茄生长的需要,又不会造成资源的浪费,符合农业可持续发展的要求。
For the loss of nutrients in tomato cultivation,the naked eye is not easy to identify the judge.To study the early identification of nutrient deficit in tomato based on computer vision technology,the study of the deficiency of nutrients in tomato was conducted using the nutrient elements of nitrogen and magnesium deficient in tomato as research object.The research images were collected by CDD camera.After the images were processed,,Extracting and characterizing the images that can show the deficit of nitrogen and magnesium.The deficient nutrition of nitrogen and magnesium in tomato can make tomato leaves behave abnormally in color and texture,so the color and texture features are extracted from the processed image,The feature images are optimized by genetic algorithm and the optimized features are combined and analyzed to establish the feature model and determine the feature vector for the analysis of the extracted feature parameters.The feature model is established by using binary tree to identify the deficiency of tomato.Based on the computer vision,the tomato deficiency identification system was simulated.The results showed that:for the tomato planting,using computer vision technology to identify the nutrient elements deficient in nitrogen and magnesium,the recognition accuracy can meet the production needs,The results of the tomato fertilization difference,both to meet the needs of tomato growth,there will be no waste of resources.And improper fertilization easily lead to environmental pollution,control the amount of fertilizer application,in line with the requirements of sustainable development.
For the loss of nutrients in tomato cultivation,the naked eye is not easy to identify the judge.To study the early identification of nutrient deficit in tomato based on computer vision technology,the study of the deficiency of nutrients in tomato was conducted using the nutrient elements of nitrogen and magnesium deficient in tomato as research object.The research images were collected by CDD camera.After the images were processed,,Extracting and characterizing the images that can show the deficit of nitrogen and magnesium.The deficient nutrition of nitrogen and magnesium in tomato can make tomato leaves behave abnormally in color and texture,so the color and texture features are extracted from the processed image,The feature images are optimized by genetic algorithm and the optimized features are combined and analyzed to establish the feature model and determine the feature vector for the analysis of the extracted feature parameters.The feature model is established by using binary tree to identify the deficiency of tomato.Based on the computer vision,the tomato deficiency identification system was simulated.The results showed that:for the tomato planting,using computer vision technology to identify the nutrient elements deficient in nitrogen and magnesium,the recognition accuracy can meet the production needs,The results of the tomato fertilization difference,both to meet the needs of tomato growth,there will be no waste of resources.And improper fertilization easily lead to environmental pollution,control the amount of fertilizer application,in line with the requirements of sustainable development.
引文
[1]应义斌,景寒松.机器视觉技术在黄花梨尺寸和果面缺陷检测中的应用[J].农业工程学报,1999,15(1):197-200.
[2]Seginer,Elster R T,Goodrum JW,et al.Plant wilt detection bycomputer-vision trackin of leaf tips[J].Trans of the ASAE,1992,35(5):1563-1567.
[3]Ling P,Cheng Z,Musacchio D J.Quantification of coffee embryo growth using machine vision[J].Trans of the ASAE,1995,38(6):1911-1917.
[4]Reid JF.Evaluation of color representations for maize image[J].Journal of Agricultural Engineering Research,1996,63:185-196.
[5]Shimizu H,Heins R D.Computer-vision-based system for plant growth analysis[J].Trans of the ASAE,1995,38(3):959-964.
[6]陈佳娟.基于图像处理和人工智能的植物病害自动诊断技术的研究[D].长春:吉林工业大学,2001.
[7]张伟.基于计算机图像处理的黄瓜缺素判别的研究[D].镇江:江苏大学,2003.
[8]赵杰文,周晓兵.基于近红外图像纹理分析检测作物叶片含水率的研究[J].农业工程学报,1999,15(3):39-42.
[9]毛罕平,吴雪梅,李萍萍.基于计算机视觉的番茄缺素神经网络识别[J].农业工程学报,2005,21(8):106-109.
[10]张彦娥,李民赞.基于计算机视觉技术的温室黄瓜叶片营养信息检测[J].农业工程学报,2005,21(8):102-105.
[11]程正兴,林勇平.小波分析在图像处理中的应用[J].工程数学学报,200l,18(5):57-86.
[12]连兆煌.无土栽培原理与技术[M].北京:中国农业出版社,1992.
[2]Seginer,Elster R T,Goodrum JW,et al.Plant wilt detection bycomputer-vision trackin of leaf tips[J].Trans of the ASAE,1992,35(5):1563-1567.
[3]Ling P,Cheng Z,Musacchio D J.Quantification of coffee embryo growth using machine vision[J].Trans of the ASAE,1995,38(6):1911-1917.
[4]Reid JF.Evaluation of color representations for maize image[J].Journal of Agricultural Engineering Research,1996,63:185-196.
[5]Shimizu H,Heins R D.Computer-vision-based system for plant growth analysis[J].Trans of the ASAE,1995,38(3):959-964.
[6]陈佳娟.基于图像处理和人工智能的植物病害自动诊断技术的研究[D].长春:吉林工业大学,2001.
[7]张伟.基于计算机图像处理的黄瓜缺素判别的研究[D].镇江:江苏大学,2003.
[8]赵杰文,周晓兵.基于近红外图像纹理分析检测作物叶片含水率的研究[J].农业工程学报,1999,15(3):39-42.
[9]毛罕平,吴雪梅,李萍萍.基于计算机视觉的番茄缺素神经网络识别[J].农业工程学报,2005,21(8):106-109.
[10]张彦娥,李民赞.基于计算机视觉技术的温室黄瓜叶片营养信息检测[J].农业工程学报,2005,21(8):102-105.
[11]程正兴,林勇平.小波分析在图像处理中的应用[J].工程数学学报,200l,18(5):57-86.
[12]连兆煌.无土栽培原理与技术[M].北京:中国农业出版社,1992.