本文選題：敲擊響應(yīng)信號(hào) + 成熟度��； 參考：《華中農(nóng)業(yè)大學(xué)》2017年碩士論文

【摘要】：西瓜成熟度無(wú)損檢測(cè)是西瓜品質(zhì)檢測(cè)研究中的一個(gè)核心問(wèn)題,近紅外、核磁共振、X射線、聲學(xué)等眾多的方法被應(yīng)用到西瓜成熟度檢測(cè)。目前西瓜品質(zhì)智能檢測(cè)技術(shù)尚未在實(shí)踐中推廣,主要原因是檢測(cè)算法的有效性、快速性、穩(wěn)定性和易實(shí)現(xiàn)性等未達(dá)到應(yīng)用上的要求。本文考慮未熟、成熟和過(guò)熟三個(gè)成熟度等級(jí)的西瓜聲學(xué)分類(lèi)問(wèn)題,通過(guò)主成分分析(Principal component analysis,PCA)和核主成分分析(kernel principal component analysis,KPCA)提取西瓜敲擊響應(yīng)信號(hào)的特征,利用極限學(xué)習(xí)機(jī)(Extreme Learning Machine,ELM)構(gòu)建西瓜成熟度分類(lèi)和西瓜含糖量檢測(cè)模型,主要的工作如下:1.提出基于(核)主成分分析的西瓜敲擊響應(yīng)信號(hào)的特征提取方法。通過(guò)不同成熟水平聲信號(hào)樣本的(核)主成分分析獲得感興趣的主成分,將其對(duì)應(yīng)的特征向量線性擴(kuò)張成有限維的特征空間,用信號(hào)在特征空間上的正交投影系數(shù)作為特征。PCA能夠在保留足夠多原始信息的同時(shí)大幅度降維,計(jì)算直接且簡(jiǎn)單。在保留95%主成分的條件下,PCA和KPCA分別將原始信號(hào)維數(shù)由4096降低到31和90。2.提出基于Markov chain采樣的極限學(xué)習(xí)機(jī)算法。ELM是一個(gè)單隱層前饋神經(jīng)網(wǎng)絡(luò)學(xué)習(xí)系統(tǒng),其輸入層權(quán)重和隱層閾值是隨機(jī)固定的,算法的核心是計(jì)算輸出矩陣的Moore-Penrose廣義逆。該算法簡(jiǎn)單、運(yùn)行速度快,且優(yōu)于支持向量機(jī)(SVM)、BP神經(jīng)網(wǎng)絡(luò)等機(jī)器學(xué)習(xí)算法。本文在Tikhonov正則化理論分析基礎(chǔ)上,將獨(dú)立同分布數(shù)據(jù)下的ELM泛化性能的進(jìn)行推廣,給出了基于Markov chain采樣的ELM算法誤差上界的估計(jì)。同時(shí),基于真實(shí)數(shù)據(jù),將獨(dú)立同分布與Markov chain采樣條件下的ELM進(jìn)行了對(duì)比分析。實(shí)驗(yàn)結(jié)果表明,Markov chain采樣不僅能有效降低ELM的預(yù)測(cè)誤差,且能夠提高ELM的魯棒性。3.構(gòu)建基于高斯核函數(shù)的KPCA-ELM西瓜成熟度分類(lèi)模型。首先,按2:1的比例將樣本進(jìn)行隨機(jī)劃分,獲得訓(xùn)練集樣本180個(gè),測(cè)試集樣本90個(gè)。其次,以ELM為分類(lèi)模型,分析了由PCA和KPCA提取的特征對(duì)西瓜成熟度檢測(cè)的影響。同時(shí),討論了基于線性、多項(xiàng)式、高斯、S型四種核函數(shù)的KPCA-ELM分類(lèi)模型對(duì)西瓜成熟度的檢測(cè)效果。最后,從分類(lèi)準(zhǔn)確率和速率兩個(gè)角度,將ELM與K-最近鄰(KNN)、BP神經(jīng)網(wǎng)絡(luò)和SVM三種分類(lèi)模型進(jìn)行了對(duì)比分析。實(shí)驗(yàn)結(jié)果表明,基于高斯核函數(shù)的KPCA-ELM的模型效果最優(yōu),在二分類(lèi)和三分類(lèi)兩種西瓜成熟度檢測(cè)場(chǎng)景下的識(shí)別準(zhǔn)確率分別為95.72%和89.23%。4.構(gòu)建基于高斯核函數(shù)的KPCA-ELM西瓜含糖量檢測(cè)模型。利用ELM構(gòu)建西瓜含糖量與敲擊響應(yīng)信號(hào)之間的回歸模型,分析了PCA、KPCA提取的特征對(duì)西瓜含糖量檢測(cè)模型的影響。同時(shí),也將ELM與偏最小二乘(PLS)、BP神經(jīng)網(wǎng)絡(luò)、支持向量回歸(SVR)進(jìn)行了對(duì)比分析。實(shí)驗(yàn)結(jié)果表明,基于KPCA-ELM的西瓜含糖量檢測(cè)模型的性能最佳,其能夠得到最小均方根誤差(Root mean square error,RMSE)為0.3725,標(biāo)準(zhǔn)偏差(Standard deviation,STD)為0.0173。
[Abstract]:Non-destructive testing of watermelon maturity is a core problem in watermelon quality testing. Many methods, such as near-infrared, nuclear magnetic resonance X-ray, acoustics and so on, have been applied to watermelon maturity detection. At present, the intelligent detection technology of watermelon quality has not been popularized in practice, the main reason is that the validity, rapidity, stability and realizability of the detection algorithm do not meet the requirements of application. In this paper, the acoustic classification of watermelon with immature, mature and overripe grades was considered. Principal component analysis (PCA) and kernel principal component analysis (KPA) were used to extract the characteristics of the knock-response signals of watermelon by principal component analysis (PCA) and kernel principal component analysis (KPA). Using extreme Learning machine (ELM) to construct the watermelon maturity classification and sugar content detection model, the main work is as follows: 1. A method for feature extraction of watermelon knock response signal based on kernel principal component analysis (PCA) is proposed. The principal components of interest are obtained by kernel principal component analysis (PCA) of different mature horizontal acoustic signal samples, and the corresponding eigenvector is linearly expanded into a finite dimensional feature space. Using the orthogonal projection coefficient of the signal in the feature space as the feature. PCA can greatly reduce the dimension while retaining enough original information. The calculation is direct and simple. Under the condition of keeping 95% principal component, the original signal dimension was reduced from 4096 to 31 and 90.2 by KPCA, respectively. A learning algorithm based on Markov chain sampling is proposed. Elm is a single hidden layer feedforward neural network learning system. Its input layer weight and hidden layer threshold are randomly fixed. The core of the algorithm is to calculate the Moore-Penrose generalized inverse of the output matrix. The algorithm is simple, fast and superior to the support vector machine (SVM) SVM / BP neural network. Based on the analysis of Tikhonov regularization theory, this paper generalizes the generalization of the ELM generalization performance under the independent same distribution data, and gives the estimation of the upper bound of the error of the ELM algorithm based on the Markov chain sampling. At the same time, based on the real data, the ELM under the condition of Markov chain sampling and independent same distribution are compared and analyzed. The experimental results show that chain sampling can not only effectively reduce the prediction error of ELM, but also improve the robustness of ELM. The maturity classification model of KPCA-ELM watermelon based on Gao Si kernel function was constructed. First, the samples are randomly divided according to 2:1 scale, and 180 samples of training set and 90 samples of test set are obtained. Secondly, using ELM as the classification model, the effects of the characteristics extracted by PCA and KPCA on watermelon maturity were analyzed. At the same time, the effect of KPCA-ELM classification model based on linear, polynomial and Gao Si S-type kernel function on watermelon maturity was discussed. Finally, from the perspective of classification accuracy and rate, three classification models, ELM, KNNNNNNBP neural network and SVM, are compared and analyzed. The experimental results show that the model of KPCA-ELM based on Gao Si kernel function is the best, and the recognition accuracy is 95.72% and 89.23.4in two kinds of watermelon maturity detection scenarios, respectively. The sugar content detection model of KPCA-ELM watermelon based on Gao Si kernel function was established. The regression model between sugar content and knock response signal of watermelon was constructed by ELM, and the influence of extraction characteristics of ELM on sugar content detection model of watermelon was analyzed. At the same time, the ELM is compared with the partial least squares BP neural network and the support vector regression (SVR). The experimental results show that the model based on KPCA-ELM has the best performance, the minimum root mean square error (RMSE) is 0.3725, and the standard deviation (STD) is 0.0173.
【學(xué)位授予單位】：華中農(nóng)業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP18;S651

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 方漢良;任志英;鄭志強(qiáng);潘輝;朱錦琳;;西瓜成熟度測(cè)試儀設(shè)計(jì)研究[J];赤峰學(xué)院學(xué)報(bào)(自然科學(xué)版);2015年12期

2 王金甲;張彥娜;;多變量自回歸和多線性主成分分析結(jié)合的多通道信號(hào)特征提取研究[J];生物醫(yī)學(xué)工程學(xué)雜志;2015年01期

3 陸勇;李臻峰;李靜;浦宏杰;徐晚秀;;聲學(xué)技術(shù)在西瓜品質(zhì)無(wú)損檢測(cè)中的應(yīng)用[J];福建農(nóng)業(yè)科技;2014年10期

4 耿榮生;景鵬;;蓬勃發(fā)展的我國(guó)無(wú)損檢測(cè)技術(shù)[J];機(jī)械工程學(xué)報(bào);2013年22期

5 鮑必賽;樓曉俊;李雋穎;劉海濤;;主成分分析在震動(dòng)信號(hào)目標(biāo)識(shí)別算法中的應(yīng)用[J];華中科技大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年07期

6 危艷君;饒秀勤;漆兵;林鵬鳥(niǎo);李江波;;基于西瓜聲學(xué)特性測(cè)定其成熟度[J];農(nóng)產(chǎn)品加工(學(xué)刊);2012年01期

7 張福軍;衣淑娟;劉坤;;西瓜成熟度無(wú)損檢測(cè)系統(tǒng)的設(shè)計(jì)[J];農(nóng)機(jī)化研究;2011年11期

8 高冠東;滕桂法;肖珂;張曉茹;;基于BMV特征的西瓜成熟度無(wú)損檢測(cè)方法[J];農(nóng)業(yè)工程學(xué)報(bào);2010年08期

9 賀艷楠;魏永勝;鄭穎;;水果成熟度無(wú)損檢測(cè)技術(shù)研究進(jìn)展[J];北方園藝;2010年03期

10 沈蓉爾;韓玉杰;吳剛;;基于MEMS與單片機(jī)的西瓜成熟度無(wú)損檢測(cè)系統(tǒng)的研究[J];機(jī)電產(chǎn)品開(kāi)發(fā)與創(chuàng)新;2010年01期

相關(guān)博士學(xué)位論文 前1條

1 楊勝凱;基于核主成分分析的特征變換研究[D];浙江大學(xué);2014年

相關(guān)碩士學(xué)位論文 前2條

1 王輝;基于核主成分分析特征提取及支持向量機(jī)的人臉識(shí)別應(yīng)用研究[D];合肥工業(yè)大學(xué);2006年

2 呂飛玲;基于聲學(xué)特性的西瓜品質(zhì)無(wú)損檢測(cè)方法研究[D];浙江大學(xué);2003年

，

本文編號(hào)：1816032