基于主成分分析的多傳感器目標(biāo)識(shí)別技術(shù)研究
本文選題:目標(biāo)識(shí)別 + LBP; 參考:《沈陽理工大學(xué)》2017年碩士論文
【摘要】:目標(biāo)識(shí)別是一個(gè)綜合了圖像處理、機(jī)器學(xué)習(xí)和模式識(shí)別等領(lǐng)域的課題,是近幾年的一個(gè)研究熱點(diǎn)。本文主要是對(duì)基于主成分分析的多傳感器目標(biāo)識(shí)別技術(shù)進(jìn)行研究,為了提高多傳感器的目標(biāo)識(shí)別率,提出了一種基于LBP-PCA的多傳感器目標(biāo)識(shí)別算法。主要研究內(nèi)容如下:首先,本文對(duì)目標(biāo)識(shí)別中的圖像預(yù)處理以及圖像描述進(jìn)行了研究。圖像預(yù)處理相關(guān)方法中重點(diǎn)研究了圖像濾波、增強(qiáng)以及分割等相關(guān)環(huán)節(jié)。在圖像濾波及增強(qiáng)環(huán)節(jié)中本文選擇的是中值濾波和直方圖均值化算法,并用實(shí)驗(yàn)檢驗(yàn)了這兩種算法對(duì)本文采集到的圖像的適用性;圖像分割環(huán)節(jié)主要包括一階微分算子以及閾值分割算法,本文對(duì)多種分割算法進(jìn)行分析和研究,通過實(shí)驗(yàn)效果的比較選取Otsu閾值分割作為本文的分割算法。圖像描述部分主要研究了各種特征提取算法,并分析了各種算法的優(yōu)缺點(diǎn),其中主要包括了圖像的顏色、形狀、紋理等特征提取方法。其中,對(duì)本文研究的LBP紋理特征進(jìn)行了簡要說明,具體研究過程放在后面的章節(jié)。對(duì)本文后續(xù)對(duì)比實(shí)驗(yàn)用到的SIFT描述子進(jìn)行研究,證明其適用于本文提出的算法,能夠有效的進(jìn)行特征描述。其次,本文選取支持向量機(jī)用于決策判斷。比較了幾種分類器的優(yōu)缺點(diǎn)。主要研究了支持向量機(jī)的分類原理,包括線性可分支持向量機(jī)、線性不可分支持向量機(jī)、非線性支持向量機(jī)及其核函數(shù)的選擇和支持向量機(jī)的應(yīng)用與拓展。再次,對(duì)基于主成分分析的異類傳感器融合算法進(jìn)行研究。分析了三種融合方式,對(duì)三種融合方式的優(yōu)缺點(diǎn)進(jìn)行比較并結(jié)合本文算法的要求,選取出特征級(jí)信息融合方式。在傳感器類型的選擇上,結(jié)合實(shí)際情況以及對(duì)各種傳感器的適用范圍選取紅外傳感器和可見光傳感器作為本文研究對(duì)象。重點(diǎn)對(duì)提出的主成分分析融合算法的各個(gè)方面進(jìn)行研究,主要包括PCA的定義、基本原理以及推導(dǎo)過程。并且通過與傳統(tǒng)串聯(lián)法以及單一傳感器的目標(biāo)識(shí)別實(shí)驗(yàn)進(jìn)行對(duì)比,可以驗(yàn)證出PCA算法可以有效地降低特征向量的維度,減少運(yùn)算量的同時(shí)保留原始數(shù)據(jù)的主要信息。最后,本文提出了一種基于LBP-PCA的多傳感器目標(biāo)識(shí)別算法,并進(jìn)行了實(shí)驗(yàn)驗(yàn)證,首先將紅外以及可見光圖像分別通過相應(yīng)的預(yù)處理算法從圖像中提取出目標(biāo),然后提取目標(biāo)的LBP特征點(diǎn)向量,再利用PCA算法降低提取出的LBP特征向量的維數(shù)從而得出融合后的特征向量,最后利用SVM進(jìn)行識(shí)別與分類用以獲取識(shí)別率。實(shí)驗(yàn)仿真結(jié)果表明LBP特征提取方法具有良好的旋轉(zhuǎn)不變性和灰度不變性,同時(shí)主成分分析可以從一個(gè)高維空間中的提取主要的特征,利用LBP-PCA多傳感器目標(biāo)識(shí)別算法可以克服傳統(tǒng)圖像融合中數(shù)據(jù)過大、運(yùn)行時(shí)間過長等問題,在實(shí)現(xiàn)實(shí)時(shí)檢測(cè)的同時(shí)提高了目標(biāo)的識(shí)別率。
[Abstract]:Target recognition is a research hotspot in recent years, which integrates the fields of image processing, machine learning and pattern recognition. In order to improve the rate of multi-sensor target recognition, a multi-sensor target recognition algorithm based on LBP-PCA is proposed in this paper. The main contents are as follows: firstly, the image preprocessing and image description in target recognition are studied in this paper. In the correlation method of image preprocessing, image filtering, image enhancement and image segmentation are studied. The median filter and histogram mean algorithm are selected in image filtering and enhancement. The applicability of these two algorithms to the images collected in this paper is verified by experiments. Image segmentation mainly includes first order differential operator and threshold segmentation algorithm. This paper analyzes and studies many segmentation algorithms. Otsu threshold segmentation is selected as the segmentation algorithm by comparing the experimental results. In the part of image description, various feature extraction algorithms are studied, and their advantages and disadvantages are analyzed, including color, shape, texture and other feature extraction methods. Among them, the LBP texture features studied in this paper are briefly described, and the specific research process is put in the following chapter. The SIFT descriptors used in the subsequent comparative experiments in this paper are studied, and it is proved that the proposed algorithm can effectively describe the features. Secondly, support vector machine is selected for decision-making. The advantages and disadvantages of several classifiers are compared. This paper mainly studies the classification principle of support vector machine, including linear separable support vector machine, linear non-separable support vector machine, nonlinear support vector machine and its kernel function, and the application and development of support vector machine. Thirdly, the fusion algorithm of heterogeneous sensors based on principal component analysis (PCA) is studied. The advantages and disadvantages of the three fusion methods are analyzed and the feature level information fusion method is selected according to the requirements of this algorithm. In the selection of sensor type, the infrared sensor and the visible light sensor are selected as the research object of this paper, combined with the actual situation and the applicable range of various sensors. All aspects of the proposed principal component analysis (PCA) fusion algorithm are studied, including the definition, the basic principle and the derivation process of PCA. By comparing with the traditional series method and the target recognition experiment of a single sensor, it can be verified that PCA algorithm can effectively reduce the dimension of the eigenvector, reduce the computation amount and retain the main information of the original data. Finally, a multi-sensor target recognition algorithm based on LBP-PCA is proposed and verified by experiments. Firstly, infrared and visible images are extracted from the image by corresponding preprocessing algorithms. Then the LBP feature point vector of the target is extracted, then the dimension of the extracted LBP feature vector is reduced by PCA algorithm, and the fused feature vector is obtained. Finally, SVM is used for recognition and classification to obtain the recognition rate. Experimental results show that LBP feature extraction method has good rotation invariance and gray invariance, and principal component analysis can extract the main features from a high-dimensional space. The LBP-PCA multi-sensor target recognition algorithm can overcome the problems of too large data and long running time in traditional image fusion. It can achieve real-time detection and improve the target recognition rate.
【學(xué)位授予單位】:沈陽理工大學(xué)
【學(xué)位級(jí)別】:碩士
【學(xué)位授予年份】:2017
【分類號(hào)】:TP391.41;TP212
【參考文獻(xiàn)】
相關(guān)期刊論文 前10條
1 侯穎妮;楊予昊;李士國;江濤;;基于ISAR像的艦船目標(biāo)識(shí)別技術(shù)研究[J];現(xiàn)代雷達(dá);2016年03期
2 倪增超;王婧;;自動(dòng)指紋識(shí)別技術(shù)的發(fā)展及應(yīng)用[J];科技創(chuàng)新與應(yīng)用;2015年32期
3 于振海;張珍明;陳茹;;基于Contourlet變換的紅外與可見光圖像融合方法[J];無線電工程;2015年08期
4 蔡磊涵;;CMOS圖像傳感器在監(jiān)控市場(chǎng)主導(dǎo)地位提升[J];中國安防;2015年10期
5 王雁;穆春陽;馬行;;基于顏色標(biāo)準(zhǔn)化模型和HOG特征的交通標(biāo)志檢測(cè)[J];軟件導(dǎo)刊;2015年03期
6 劉小楠;馬彩霞;劉峰;郎海濤;;基于光學(xué)顯微圖像特征的羊絨識(shí)別技術(shù)[J];毛紡科技;2014年10期
7 李新德;潘錦東;DEZERT Jean;;一種基于DSmT和HMM的序列飛機(jī)目標(biāo)識(shí)別算法[J];自動(dòng)化學(xué)報(bào);2014年12期
8 張成軍;陰妍;鮑久圣;紀(jì)洋洋;;多源信息融合故障診斷方法研究進(jìn)展[J];河北科技大學(xué)學(xué)報(bào);2014年03期
9 李小娟;席曉燕;臧義華;梁佳;;基于粒子群優(yōu)化的圖像邊緣融合算法[J];軟件導(dǎo)刊;2014年05期
10 吳冬梅;李俊威;劉凌志;藺麗華;;基于Zernike不變矩的人形識(shí)別研究[J];計(jì)算機(jī)應(yīng)用與軟件;2013年08期
相關(guān)博士學(xué)位論文 前1條
1 呂燕;基于動(dòng)態(tài)PLS方法的建模及預(yù)測(cè)控制器設(shè)計(jì)[D];浙江大學(xué);2013年
相關(guān)碩士學(xué)位論文 前10條
1 方晟;基于人臉特征的性別判斷和年齡估計(jì)方法研究[D];浙江大學(xué);2015年
2 齊光景;基于fast-AdaBoost算法的人臉檢測(cè)與識(shí)別方法研究[D];太原理工大學(xué);2014年
3 鄧慧萍;基于統(tǒng)計(jì)學(xué)習(xí)的機(jī)場(chǎng)跑道異物檢測(cè)[D];電子科技大學(xué);2014年
4 史陽陽;旋翼型無人機(jī)自主著艦?zāi)繕?biāo)識(shí)別技術(shù)研究[D];南京航空航天大學(xué);2013年
5 楊陽;基于紅外與可見光圖像的特征融合方法研究[D];沈陽理工大學(xué);2013年
6 王東明;基于特征級(jí)融合的目標(biāo)識(shí)別方法研究[D];沈陽理工大學(xué);2013年
7 唐春益;AdaBoost算法及其在目標(biāo)識(shí)別中的應(yīng)用研究[D];南昌航空大學(xué);2012年
8 陳建;基于腦出血CT圖像的分割與提取算法研究[D];安徽大學(xué);2012年
9 劉麗霞;圖像紋理特征研究和比較[D];北京郵電大學(xué);2011年
10 李樹娟;基于LBP特征的人臉表情分析[D];中國石油大學(xué);2010年
,本文編號(hào):2053145
本文鏈接:http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2053145.html