【摘要】：隨著葡萄市場(chǎng)經(jīng)濟(jì)的發(fā)展,葡萄品種識(shí)別對(duì)于葡萄這種經(jīng)濟(jì)作物的科普和市場(chǎng)推廣都具有重要意義。在葡萄品種識(shí)別研究中,一般以葉片為識(shí)別研究對(duì)象,主要考慮到葉片相對(duì)于果實(shí)來說易于保存且可采摘時(shí)間長(zhǎng),研究過程中不需要其他學(xué)科的輔助實(shí)驗(yàn)。但是葡萄葉片識(shí)別存在一個(gè)明顯的難點(diǎn),這種同科屬類葉片顏色和形態(tài)結(jié)構(gòu)差異小,使得識(shí)別研究中準(zhǔn)確率不高。為解決這個(gè)問題,本論文展開了針對(duì)同屬類葡萄葉片識(shí)別的研究,提出一種基于流形學(xué)習(xí)的葡萄葉片品種識(shí)別方案。本研究采用了15個(gè)品種的450幅葡萄葉片作為實(shí)驗(yàn)樣本,進(jìn)行葡萄品種的分類識(shí)別實(shí)驗(yàn)。重點(diǎn)研究葡萄葉片的特征提取和特征降維。在特征提取部分,分別采用人工設(shè)計(jì)特征的灰度共生矩陣、方向梯度直方圖、可變性部件模型和由卷積神經(jīng)網(wǎng)絡(luò)提取的深度學(xué)習(xí)特征作為葉片特征,分析特征數(shù)據(jù)性質(zhì)和表現(xiàn)能力。發(fā)現(xiàn)高維度的特征表示葡萄葉片的能力優(yōu)于低維度,但是高維特征數(shù)據(jù)量大,冗余性高,雖然能夠得到較好的識(shí)別結(jié)果,但效率較低。為了降低高維葡萄葉片特征數(shù)據(jù)在識(shí)別過程中的復(fù)雜度,提高其實(shí)用性與實(shí)驗(yàn)效率,本文采用流形學(xué)習(xí)算法對(duì)提取的高維葡萄葉片特征進(jìn)行降維,在保持識(shí)別精度的基礎(chǔ)上,提高算法的效率,使其具有實(shí)用性。在葡萄葉片特征降維研究中,分別應(yīng)用了局部線性嵌入(LLE)、拉普拉斯特征映射(LE)、局部保持投影(LPP)、臨近保持嵌入(NPE)四種不同的算法進(jìn)行特征降維,得到葡萄葉片在低維空間的特征表示,并對(duì)影響降維性能的重點(diǎn)參數(shù)進(jìn)行了分析;在葡萄葉片識(shí)別過程中,對(duì)比分析不同分類器的分類效果,最后通過訓(xùn)練支持向量機(jī)(SVM)分類模型,進(jìn)行葉片分類識(shí)別。本論文通過實(shí)驗(yàn)分析驗(yàn)證了流形降維在葡萄葉片識(shí)別中的可行性和必要性,流形降維能夠有效保持?jǐn)?shù)據(jù)在高維空間的內(nèi)部結(jié)構(gòu)特征。降維后的特征,在提高識(shí)別速度的同時(shí),相對(duì)于降維前,仍具有良好的葉片識(shí)別準(zhǔn)確性。其中利用卷積神經(jīng)網(wǎng)絡(luò)進(jìn)行特征提取結(jié)合流形學(xué)習(xí)算法對(duì)其進(jìn)行降維,識(shí)別率最高可以達(dá)到90.33%,識(shí)別性能優(yōu)于降維前的性能,并且識(shí)別速度大幅提高,相較于不做特征降維的識(shí)別時(shí)間而言,時(shí)間縮短為原來的1/3。對(duì)于人工設(shè)計(jì)特征的降維,降維后的識(shí)別時(shí)間有明顯的改善,其中DPM特征維度降為原來的1/30時(shí),識(shí)別用時(shí)縮短為原來的1/6。本文的研究為葡萄葉片的快速識(shí)別,提供了一種有效的方法。
[Abstract]:With the development of grape market economy, grape variety identification is of great significance to the popularization of science and marketing of this kind of cash crop. In the research of grape variety recognition, the leaf is generally regarded as the object of study, considering that the leaf is easy to be preserved and can be picked for a long time compared with the fruit, and there is no need for the auxiliary experiment of other disciplines in the course of the research. However, there is an obvious difficulty in the leaf recognition of grape. The difference of leaf color and morphological structure of the same family, genus and species is small, which makes the accuracy of the recognition research not high. In order to solve this problem, this paper studies the leaf recognition of the same genus grape, and proposes a new method of grape leaf variety recognition based on manifold learning. In this study, 450 leaves of 15 grape varieties were used as experimental samples to classify and identify grape varieties. The feature extraction and dimensionality reduction of grape leaves were studied. In the part of feature extraction, the grayscale co-occurrence matrix, directional gradient histogram, variable component model and depth learning feature extracted from convolutional neural network are used as leaf features, respectively. Analyze the nature and performance of the feature data. It was found that the high dimension feature indicates that the ability of grape leaves is superior to that of the low dimension, but the high dimensional feature has large data volume and high redundancy, which can obtain better recognition results, but its efficiency is low. In order to reduce the complexity of high dimensional grape leaf feature data in recognition process and improve its practicability and experimental efficiency, this paper adopts manifold learning algorithm to reduce the dimension of the extracted high dimensional grape leaf feature, on the basis of maintaining the recognition accuracy. Improve the efficiency of the algorithm and make it practical. In the study of grape leaf feature dimensionality reduction, four different algorithms of locally linear embedded (LLE), Laplacian feature map, (LE), local preserving projection, (LPP), near preserving embedding (NPE), were used to reduce the dimension of grape leaves. The characteristic representation of grape leaves in low-dimensional space was obtained, and the key parameters affecting the performance of reducing dimension were analyzed. In the process of grape leaf recognition, the classification effect of different classifiers was compared and analyzed. Finally, by training support vector machine (SVM) classification model, the blade classification recognition. In this paper, the feasibility and necessity of manifold dimensionality reduction in grape blade recognition are verified by experimental analysis. Manifold dimensionality reduction can effectively maintain the internal structural characteristics of data in high-dimensional space. The features after dimensionality reduction can improve the recognition speed and still have good accuracy of blade recognition compared with those before dimensionality reduction. Among them, convolution neural network is used for feature extraction and manifold learning algorithm to reduce the dimension, the recognition rate can reach 90.33, the recognition performance is better than that before dimensionality reduction, and the recognition speed is greatly improved. Compared with the recognition time without feature reduction, the time is shortened to one third of the original time. For the dimensionality reduction of artificial design features, the recognition time after dimensionality reduction is obviously improved. When the DPM feature dimension is reduced to 1 / 30 of the original, the recognition time is shortened to 1 / 6 of the original. The research in this paper provides an effective method for rapid recognition of grape leaves.
【學(xué)位授予單位】：西北農(nóng)林科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：S663.1;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 董冀媛;曾慧;穆志純;付冬梅;;局部切空間排列多姿態(tài)人耳識(shí)別[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2015年05期

2 馬媛;馮全;楊梅;李妙祺;;基于HOG的釀酒葡萄葉檢測(cè)[J];計(jì)算機(jī)工程與應(yīng)用;2016年15期

3 黃宏臣;韓振南;張倩倩;李月仙;張志偉;;基于拉普拉斯特征映射的滾動(dòng)軸承故障識(shí)別[J];振動(dòng)與沖擊;2015年05期

4 任國(guó)貞;江濤;;基于灰度共生矩陣的紋理提取方法研究[J];計(jì)算機(jī)應(yīng)用與軟件;2014年11期

5 王偉;王秀蘭;馮仲科;劉芳;;基于Android手機(jī)的樹木葉片識(shí)別系統(tǒng)[J];廣東農(nóng)業(yè)科學(xué);2014年18期

6 季偉;王力;;基于流形學(xué)習(xí)LPP算法的語(yǔ)音特征提取應(yīng)用[J];通信技術(shù);2013年12期

7 賈淵;李振江;彭增起;;結(jié)合LLE流形學(xué)習(xí)和支持向量機(jī)的豬肉顏色分級(jí)[J];農(nóng)業(yè)工程學(xué)報(bào);2012年09期

8 羅芳瓊;;LLE流形學(xué)習(xí)的若干問題分析[J];現(xiàn)代計(jì)算機(jī)(專業(yè)版);2012年08期

9 姚明海;瞿心昱;;基于自適應(yīng)子空間在線PCA的手勢(shì)識(shí)別[J];模式識(shí)別與人工智能;2011年02期

10 湯曉東;劉滿華;趙輝;陶衛(wèi);;復(fù)雜背景下的大豆葉片識(shí)別[J];電子測(cè)量與儀器學(xué)報(bào);2010年04期

相關(guān)會(huì)議論文 前1條

1 曾九孫;郜傳厚;羅世華;李啟會(huì);;基于增量LPP的在線過程監(jiān)控方法及其應(yīng)用[A];中國(guó)自動(dòng)化學(xué)會(huì)控制理論專業(yè)委員會(huì)C卷[C];2011年

相關(guān)博士學(xué)位論文 前4條

1 雷迎科;流形學(xué)習(xí)算法及其應(yīng)用研究[D];中國(guó)科學(xué)技術(shù)大學(xué);2011年

2 王燦;基于半監(jiān)督流形學(xué)習(xí)的Web信息檢索技術(shù)研究[D];浙江大學(xué);2009年

3 李波;基于流形學(xué)習(xí)的特征提取方法及其應(yīng)用研究[D];中國(guó)科學(xué)技術(shù)大學(xué);2008年

4 黃啟宏;流形學(xué)習(xí)方法理論研究及圖像中應(yīng)用[D];電子科技大學(xué);2007年

相關(guān)碩士學(xué)位論文 前3條

1 孫宏杰;基于葉片圖像分析的葡萄品種識(shí)別方法研究[D];西北農(nóng)林科技大學(xué);2016年

2 楊曉靜;基于流形學(xué)習(xí)的數(shù)據(jù)聚類與可視化[D];西安電子科技大學(xué);2013年

3 劉俊寧;基于LPP算法的人臉識(shí)別技術(shù)研究[D];江蘇大學(xué);2010年

，

本文編號(hào)：2180731