基于高光譜信息的柑橘葉綠素含量預(yù)測(cè)模型研究
發(fā)布時(shí)間：2018-11-13 16:34
【摘要】：柑橘是世界第一大果樹作物,我國(guó)的柑橘產(chǎn)量和面積均已位居世界首位。柑橘產(chǎn)業(yè)已成為我國(guó)南方果農(nóng)的主要經(jīng)濟(jì)來(lái)源。通過(guò)對(duì)柑橘葉綠素含量的分析能準(zhǔn)確掌握果樹的光合能力、營(yíng)養(yǎng)狀況和生長(zhǎng)態(tài)勢(shì),為果園管理提供科學(xué)指導(dǎo)。測(cè)定葉綠素含量的傳統(tǒng)方法是分光光度法,利用化學(xué)試劑萃取葉片中的葉綠素,依據(jù)不同波長(zhǎng)下葉綠素吸光度不同的原理計(jì)算得到葉綠素含量。這種檢測(cè)手段耗時(shí)長(zhǎng),具有破壞性,同時(shí)還依賴于檢測(cè)者的操作技術(shù),無(wú)法在數(shù)字化農(nóng)業(yè)中推廣。隨著高光譜技術(shù)和遙感技術(shù)的發(fā)展,基于光譜信息建立預(yù)測(cè)模型已成為作物估產(chǎn)和營(yíng)養(yǎng)檢測(cè)的新手段。這種方法依據(jù)的是物質(zhì)固有的吸收、發(fā)射或散射光譜特性。與傳統(tǒng)化學(xué)分析手段相比,具有無(wú)損、快捷、準(zhǔn)確的優(yōu)點(diǎn),符合現(xiàn)代農(nóng)業(yè)的發(fā)展要求。目前基于遙感技術(shù)的營(yíng)養(yǎng)診斷主要應(yīng)用在玉米、水稻等大田作物上,對(duì)柑橘等單株植物的研究相對(duì)較少。建立預(yù)測(cè)模型的常用方法有支持向量回歸、BP神經(jīng)網(wǎng)絡(luò)、多元線性回歸等,但涉及BP神經(jīng)網(wǎng)絡(luò)優(yōu)化技術(shù)的研究卻很少。本文以12年生枳橙[C sinensis(L.)Osbeck×P.trifoliate(L.)Raf.'Carrizo citrage']砧紐荷爾臍橙(C sinensis(L.)Osbeck'Newhall navel orange')為研究對(duì)象,重點(diǎn)研究粒子群優(yōu)化神經(jīng)網(wǎng)絡(luò)算法,并利用高光譜信息在葉片級(jí)別建立葉綠素含量預(yù)測(cè)模型。旨在提高葉綠素含量預(yù)測(cè)精度,同時(shí)為高光譜遙感在柑橘長(zhǎng)勢(shì)監(jiān)測(cè)中的應(yīng)用提供理論依據(jù)和技術(shù)支持。本文的主要研究?jī)?nèi)容可歸納為以下兩個(gè)方面:(1)提出了一種改進(jìn)的粒子群優(yōu)化算法,并用來(lái)優(yōu)化BP神經(jīng)網(wǎng)絡(luò)。傳統(tǒng)的BP算法在訓(xùn)練網(wǎng)絡(luò)時(shí)容易陷入局部最優(yōu),為解決這一問(wèn)題,研究者們提出了多種優(yōu)化算法。研究表明,這些優(yōu)化技術(shù)能有效提高BP神經(jīng)網(wǎng)絡(luò)的性能,已在多個(gè)領(lǐng)域得到了應(yīng)用。本文分析了以往葉綠素含量檢測(cè)研究中常用的建模方法,發(fā)現(xiàn)較少涉及BP神經(jīng)網(wǎng)絡(luò)的優(yōu)化算法。為了進(jìn)一步提高預(yù)測(cè)模型的準(zhǔn)確性,本文將粒子群優(yōu)化BP神經(jīng)網(wǎng)絡(luò)算法作為研究重點(diǎn)。針對(duì)粒子群算法中適應(yīng)度信息未被充分利用的問(wèn)題,在原有算法的基礎(chǔ)上提出了一種改進(jìn)的粒子群優(yōu)化算法(FDPSOs),并用它替代原有粒子群算法對(duì)BP神經(jīng)網(wǎng)絡(luò)權(quán)值進(jìn)行優(yōu)化。為了驗(yàn)證算法的性能,本文選擇四組分類數(shù)據(jù)集進(jìn)行實(shí)驗(yàn),并與BP及其他多種改進(jìn)的PSOs-BP網(wǎng)絡(luò)進(jìn)行比較。實(shí)驗(yàn)結(jié)果表明,算法能有效優(yōu)化BP神經(jīng)網(wǎng)絡(luò),提高模型的學(xué)習(xí)和泛化性能。(2)建立基于光譜信息的柑橘葉綠素含量預(yù)測(cè)模型。本研究將采集的柑橘葉片室內(nèi)光譜轉(zhuǎn)換為一階導(dǎo)數(shù)形式、二階導(dǎo)數(shù)形式和log(1/r)形式。通過(guò)主成分分析法和連續(xù)投影法處理多種形式的光譜數(shù)據(jù),分別得到降維后的特征向量和特征波長(zhǎng)。選擇多元線性回歸、偏最小二乘法、支持向量回歸、BP神經(jīng)網(wǎng)絡(luò)以及FDPSOs優(yōu)化后的BP神經(jīng)網(wǎng)絡(luò)建立柑橘葉綠素含量預(yù)測(cè)模型。比較不同模型的預(yù)測(cè)結(jié)果得到如下結(jié)論:經(jīng)FDPSOs優(yōu)化后的BP神經(jīng)網(wǎng)絡(luò)在預(yù)測(cè)葉綠素含量時(shí)比單純的BP神經(jīng)網(wǎng)絡(luò)準(zhǔn)確性更高(R=0.8786,RMSE=0.1683);使用原始光譜數(shù)據(jù)和log(1/r)形式的光譜數(shù)據(jù)進(jìn)行柑橘葉綠素含量預(yù)測(cè)比導(dǎo)數(shù)形式的數(shù)據(jù)更有效。
[Abstract]:Citrus is the first fruit tree in the world, and the yield and area of citrus in China are the first in the world. The citrus industry has become the main economic source of the fruit farmers in the south of China. Through the analysis of the content of the chlorophyll of the citrus, the photosynthetic capacity, the nutritional status and the growth of the fruit trees can be accurately controlled, and the scientific guidance is provided for the management of the orchards. The traditional method for measuring the content of chlorophyll is to determine the content of chlorophyll from the principle of the different chlorophyll absorbance at different wavelengths by the method of spectrophotometry and the extraction of the chlorophyll in the leaves by the chemical reagent. The detection method is time-consuming and destructive, and also depends on the operator's operation technology and cannot be popularized in the digital agriculture. With the development of high-spectrum technology and remote sensing technology, the establishment of the prediction model based on the spectral information has become a new means of crop estimation and nutrition detection. This method is based on the intrinsic absorption, emission or scattering spectrum characteristics of the substance. Compared with the traditional chemical analysis method, the method has the advantages of non-destructive, rapid and accurate, and accords with the development requirement of modern agriculture. At present, the nutrition diagnosis based on remote sensing technology is mainly applied to large field crops such as corn, rice and the like, and the research on the plant of the single plant such as the citrus is relatively small. The common methods to set up the prediction model are support vector regression, BP neural network, multiple linear regression, etc., but the research of BP neural network optimization technology is very low. This paper takes 12 years of raw orange[C sinensis (L.) Osbeck and P. trifoliate (L.) Raf. 'Carrizo citrage.'] Anvil's navel orange (C sinensis (L.) Osbeck 'Newhall namel) In order to study the object, the particle swarm optimization neural network (PSO) algorithm is focused on, and the chlorophyll content prediction model is established at the blade level by using the high spectral information. The aim of this paper is to improve the accuracy of the prediction of the content of chlorophyll, and to provide the theoretical basis and technical support for the application of high-spectral remote sensing in the monitoring of the long-term citrus. The main contents of this paper can be summarized as follows: (1) An improved particle swarm optimization algorithm is proposed and used to optimize the BP neural network. The traditional BP algorithm is easy to get into the local optimal when training the network, and to solve the problem, the researchers put forward a variety of optimization algorithms. The results show that these optimization techniques can effectively improve the performance of BP neural network and have been applied in many fields. In this paper, the commonly used modeling method in the research of chlorophyll content detection is analyzed, and the optimization algorithm of BP neural network is found to be less. In order to further improve the accuracy of the prediction model, the particle swarm optimization BP neural network algorithm is used as the research focus. In order to solve the problem that the fitness information is not fully utilized in the particle swarm optimization algorithm, an improved particle swarm optimization algorithm (FDPSOs) is proposed on the basis of the original algorithm, and the BP neural network weight value is optimized by using it instead of the original particle swarm optimization algorithm. In order to verify the performance of the algorithm, four groups of classified data sets are selected and compared with BP and other modified PSOs-BP networks. The experimental results show that the algorithm can effectively optimize the BP neural network and improve the learning and generalization performance of the model. and (2) establishing a citrus chlorophyll content prediction model based on the spectral information. In this study, the indoor spectrum of citrus leaves was converted into a first derivative form, a second derivative form and a log (1/ r) form. and processing the spectral data in various forms by the principal component analysis method and the continuous projection method, and respectively obtaining the characteristic vector and the characteristic wavelength after the dimension reduction. The prediction model of the chlorophyll content of citrus was established by using the BP neural network after the multiple linear regression, the partial least square method, the support vector regression, the BP neural network and the FDPSOs optimization. The results of the prediction of different models are as follows: the BP neural network optimized by FDPSOs is more accurate than the simple BP neural network in predicting the content of chlorophyll (R = 0.8786, RMSE = 0.1683); The use of the spectral data in the form of the original spectral data and the log (1/ r) is more effective than the data in the derivative form.
【學(xué)位授予單位】：西南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：S666;TP18
本文編號(hào)：2329699
[Abstract]:Citrus is the first fruit tree in the world, and the yield and area of citrus in China are the first in the world. The citrus industry has become the main economic source of the fruit farmers in the south of China. Through the analysis of the content of the chlorophyll of the citrus, the photosynthetic capacity, the nutritional status and the growth of the fruit trees can be accurately controlled, and the scientific guidance is provided for the management of the orchards. The traditional method for measuring the content of chlorophyll is to determine the content of chlorophyll from the principle of the different chlorophyll absorbance at different wavelengths by the method of spectrophotometry and the extraction of the chlorophyll in the leaves by the chemical reagent. The detection method is time-consuming and destructive, and also depends on the operator's operation technology and cannot be popularized in the digital agriculture. With the development of high-spectrum technology and remote sensing technology, the establishment of the prediction model based on the spectral information has become a new means of crop estimation and nutrition detection. This method is based on the intrinsic absorption, emission or scattering spectrum characteristics of the substance. Compared with the traditional chemical analysis method, the method has the advantages of non-destructive, rapid and accurate, and accords with the development requirement of modern agriculture. At present, the nutrition diagnosis based on remote sensing technology is mainly applied to large field crops such as corn, rice and the like, and the research on the plant of the single plant such as the citrus is relatively small. The common methods to set up the prediction model are support vector regression, BP neural network, multiple linear regression, etc., but the research of BP neural network optimization technology is very low. This paper takes 12 years of raw orange[C sinensis (L.) Osbeck and P. trifoliate (L.) Raf. 'Carrizo citrage.'] Anvil's navel orange (C sinensis (L.) Osbeck 'Newhall namel) In order to study the object, the particle swarm optimization neural network (PSO) algorithm is focused on, and the chlorophyll content prediction model is established at the blade level by using the high spectral information. The aim of this paper is to improve the accuracy of the prediction of the content of chlorophyll, and to provide the theoretical basis and technical support for the application of high-spectral remote sensing in the monitoring of the long-term citrus. The main contents of this paper can be summarized as follows: (1) An improved particle swarm optimization algorithm is proposed and used to optimize the BP neural network. The traditional BP algorithm is easy to get into the local optimal when training the network, and to solve the problem, the researchers put forward a variety of optimization algorithms. The results show that these optimization techniques can effectively improve the performance of BP neural network and have been applied in many fields. In this paper, the commonly used modeling method in the research of chlorophyll content detection is analyzed, and the optimization algorithm of BP neural network is found to be less. In order to further improve the accuracy of the prediction model, the particle swarm optimization BP neural network algorithm is used as the research focus. In order to solve the problem that the fitness information is not fully utilized in the particle swarm optimization algorithm, an improved particle swarm optimization algorithm (FDPSOs) is proposed on the basis of the original algorithm, and the BP neural network weight value is optimized by using it instead of the original particle swarm optimization algorithm. In order to verify the performance of the algorithm, four groups of classified data sets are selected and compared with BP and other modified PSOs-BP networks. The experimental results show that the algorithm can effectively optimize the BP neural network and improve the learning and generalization performance of the model. and (2) establishing a citrus chlorophyll content prediction model based on the spectral information. In this study, the indoor spectrum of citrus leaves was converted into a first derivative form, a second derivative form and a log (1/ r) form. and processing the spectral data in various forms by the principal component analysis method and the continuous projection method, and respectively obtaining the characteristic vector and the characteristic wavelength after the dimension reduction. The prediction model of the chlorophyll content of citrus was established by using the BP neural network after the multiple linear regression, the partial least square method, the support vector regression, the BP neural network and the FDPSOs optimization. The results of the prediction of different models are as follows: the BP neural network optimized by FDPSOs is more accurate than the simple BP neural network in predicting the content of chlorophyll (R = 0.8786, RMSE = 0.1683); The use of the spectral data in the form of the original spectral data and the log (1/ r) is more effective than the data in the derivative form.
【學(xué)位授予單位】：西南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：S666;TP18
【參考文獻(xiàn)】
相關(guān)期刊論文 前10條
1 譚林;何秉宇;劉衛(wèi)國(guó);龐冬;;基于優(yōu)化SVR高光譜指數(shù)的獨(dú)尾草葉綠素含量估算[J];生態(tài)學(xué)雜志;2017年02期
2 張婉婉;楊可明;汪國(guó)平;劉二雄;劉聰;;基于EMD-SD光譜的玉米葉片葉綠素含量GA-BP模型反演[J];浙江農(nóng)業(yè)學(xué)報(bào);2016年08期
3 程志慶;張勁松;孟平;李巖泉;王鶴松;李春友;;楊樹葉片葉綠素含量高光譜估算模型研究[J];農(nóng)業(yè)機(jī)械學(xué)報(bào);2015年08期
4 李艷嬌;李瑞敏;陳經(jīng)偉;;多元線性回歸的MATLAB實(shí)現(xiàn)[J];常熟理工學(xué)院學(xué)報(bào);2014年02期
5 張景陽(yáng);潘光友;;多元線性回歸與BP神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型對(duì)比與運(yùn)用研究[J];昆明理工大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年06期
6 尹光志;李銘輝;李文璞;曹偈;李星;;基于改進(jìn)BP神經(jīng)網(wǎng)絡(luò)的煤體瓦斯?jié)B透率預(yù)測(cè)模型[J];煤炭學(xué)報(bào);2013年07期
7 林卉;梁亮;張連蓬;杜培軍;;基于支持向量機(jī)回歸算法的小麥葉面積指數(shù)高光譜遙感反演[J];農(nóng)業(yè)工程學(xué)報(bào);2013年11期
8 王素立;劉永;;基于波動(dòng)相關(guān)性及主分量變換的多元線性回歸模型研究[J];統(tǒng)計(jì)與決策;2012年22期
9 李永亮;張懷清;林輝;;基于紅邊參數(shù)與PCA的GA-BP神經(jīng)網(wǎng)絡(luò)估算葉綠素含量模型[J];林業(yè)科學(xué);2012年09期
10 歐文娟;孟耀勇;張小燕;孔猛;;紫外可見(jiàn)吸收光譜結(jié)合主成分-反向傳播人工神經(jīng)網(wǎng)絡(luò)鑒別真假蜂蜜[J];分析化學(xué);2011年07期
相關(guān)博士學(xué)位論文 前3條
1 熊巍;中國(guó)柑橘產(chǎn)銷預(yù)警系統(tǒng)構(gòu)建及應(yīng)用研究[D];華中農(nóng)業(yè)大學(xué);2015年
2 龔夢(mèng);中國(guó)柑橘鮮果價(jià)格形成及影響因素研究[D];華中農(nóng)業(yè)大學(xué);2013年
3 涂娟娟;PSO優(yōu)化神經(jīng)網(wǎng)絡(luò)算法的研究及其應(yīng)用[D];江蘇大學(xué);2013年
相關(guān)碩士學(xué)位論文 前5條
1 王卓遠(yuǎn);基于高光譜的蘋果樹葉片葉綠素與氮素含量估測(cè)[D];山東農(nóng)業(yè)大學(xué);2015年
2 陳佩;主成分分析法研究及其在特征提取中的應(yīng)用[D];陜西師范大學(xué);2014年
3 陸洪濤;偏最小二乘回歸數(shù)學(xué)模型及其算法研究[D];華北電力大學(xué);2014年
4 祝高明;基于光譜數(shù)據(jù)和農(nóng)學(xué)參數(shù)的柑橘估產(chǎn)研究[D];西南大學(xué);2012年
5 楊芳;基于支持向量回歸(SVR)的材料熱加工過(guò)程建模[D];上海交通大學(xué);2010年
，本文編號(hào)：2329699
本文鏈接：http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2329699.html
最近更新
教材專著
