發(fā)布時間：2018-04-15 22:17

  本文選題：合成孔徑雷達(dá) + 自動目標(biāo)識別　； 參考：《五邑大學(xué)》2017年碩士論文

【摘要】：圖像識別是模式識別領(lǐng)域的典型應(yīng)用,它通過計算機(jī)算法對圖像進(jìn)行處理、分析和理解,以識別圖像中各種不同模式的對象。合成孔徑雷達(dá)(Synthetic Aperture Radar,SAR)圖像在當(dāng)今科技的快速發(fā)展且日新月異的情況下,在各個領(lǐng)域尤其是軍事領(lǐng)域發(fā)揮越來越多的作用,越來越多的學(xué)者將研究重心集中在SAR圖像目標(biāo)識別。而其中的自動目標(biāo)識別(Automatic Target Recognition,ATR)技術(shù)是SAR圖像識別系統(tǒng)的核心技術(shù),由于ATR在對圖像特征的學(xué)習(xí)以及提取過程所發(fā)揮的作用,使其成為國內(nèi)外學(xué)者的研究熱點。ATR主要借助于SAR圖像數(shù)據(jù),依靠計算機(jī)算法對圖像目標(biāo)區(qū)域進(jìn)行定位,并從中提取出包含大量目標(biāo)信息的圖像特征,并對特征信息進(jìn)行識別。近年來,很多SAR圖像目標(biāo)識別算法被提出,類似于其他圖像識別算法的三個過程:圖像預(yù)處理、圖像目標(biāo)特征提取和圖像目標(biāo)識別,本文同樣從該三個方面出發(fā),主要做了以下工作:(1)對于圖像預(yù)處理,SAR圖像中有大量的背景噪聲以及圖像目標(biāo)占SAR圖像比例較小,本文首先采用質(zhì)心定位法尋找SAR圖像感興趣區(qū)域(Region of Interest,ROI),通過對原始圖像感興趣區(qū)域的提取以達(dá)到去除SAR圖像背景噪聲的效果。在本文的實驗中,SAR原始圖像中提取的ROI圖像大小分別為49×49和64×64,實驗結(jié)果表明ROI圖像可以很好地去除SAR圖像背景噪聲;其次,在機(jī)器學(xué)習(xí)尤其是無監(jiān)督學(xué)習(xí)和深度學(xué)習(xí)中,算法對訓(xùn)練數(shù)據(jù)的數(shù)據(jù)量有較高要求,同時SAR圖像往往局限于圖像數(shù)據(jù)量,因此,本文結(jié)合SAR圖像成像時對目標(biāo)方位角敏感的特點,提出了兩種基于SAR圖像的數(shù)據(jù)增強(qiáng)的方法,通過旋轉(zhuǎn)圖像中目標(biāo)物體的方位角,以及在原始圖像上增加隨機(jī)整數(shù)值來得到更多圖像數(shù)據(jù),實驗結(jié)果表明所提方法具有有效性。(2)深度學(xué)習(xí)(Deep Learning,DL)是機(jī)器學(xué)習(xí)中一大熱門,在許多領(lǐng)域如圖像識別、語音識別、自然語言處理等領(lǐng)域取得了突破。卷積神經(jīng)網(wǎng)絡(luò)(convolutional neural networks,CNN)是深度學(xué)習(xí)算法的一種,本文結(jié)合CNN原理,訓(xùn)練了一個深度CNN網(wǎng)絡(luò)模型SARnet,SARnet包含兩個卷積層,卷積核大小為7×7、兩個池化層、兩個全連接層;結(jié)合數(shù)據(jù)增強(qiáng)方法對訓(xùn)練集擴(kuò)充為原始數(shù)據(jù)庫的22倍之后,對網(wǎng)絡(luò)進(jìn)行訓(xùn)練,并用訓(xùn)練得到的網(wǎng)絡(luò)模型對MSTAR數(shù)據(jù)庫測試集進(jìn)行特征提取,用SVM對提取出的特征進(jìn)行分類,最終識別率達(dá)到了95.68%,該識別率高于其他CNN模型。(3)近來在機(jī)器學(xué)習(xí)中,研究學(xué)者已經(jīng)將目光集中在從一些沒有標(biāo)簽的數(shù)據(jù)中學(xué)習(xí)特征,本文針對SAR圖像特征提取,先采用感興趣區(qū)域提取使輸入圖像大小為64×64;然后采用無監(jiān)督K均值特征學(xué)習(xí)算法,結(jié)合數(shù)據(jù)增強(qiáng)后的訓(xùn)練數(shù)據(jù)庫,學(xué)習(xí)到一些有用的表示。通過分塊自編碼和優(yōu)化接受域參數(shù)進(jìn)行SAR圖像特征學(xué)習(xí)可以使模型學(xué)習(xí)到多樣性的特征,并通過實驗證明,結(jié)合K-means的無監(jiān)督特征學(xué)習(xí)得到的特征可以使SAR圖像識別率達(dá)到96.67%的主流識別率。
[Abstract]:Image recognition is a typical application in the field of pattern recognition, which based on computer image processing, analysis and understanding, to identify images in different modes. The synthetic aperture radar (Synthetic Aperture, Radar, SAR) image in the rapid development of science and technology change rapidly and the situation in various fields, especially play a more and more the role of the military field, more and more scholars focused on SAR image target recognition. The automatic target recognition (Automatic Target Recognition ATR) technology is the core technology of SAR image recognition system don't, because ATR plays in the image feature extraction process and the learning effect, make it become a hot research topic.ATR at home and abroad with the help of SAR image data, relying on the target region of image positioning algorithm, and extract contains a large number of orders Image character information, and identification of feature information. In recent years, a lot of SAR image target recognition algorithm is proposed. The three process is similar to the other image recognition algorithm: image preprocessing, image feature extraction and image recognition, this paper also from the three aspects, mainly do the following work: (1) for image preprocessing, SAR images have a lot of background noise and image target accounted for a smaller proportion of SAR image, this paper uses centroid positioning method for SAR image region of interest (Region of, Interest, ROI), in order to remove the SAR image background noise effect on the original image by extracting the region of interest. In this experiment, the ROI image size extraction were 49 * 49 and 64 * 64 SAR in the original image. Experimental results show that the ROI image can remove the background noise of SAR image; secondly, especially in machine learning It is an unsupervised learning and deep learning, data algorithm based on the amount of training data have higher requirements, at the same time, SAR images are often limited to the amount of image data, therefore, this paper combines the SAR image imaging features of target azimuth sensitivity, puts forward two methods for image enhancement based on SAR data, through the object rotation the image in azimuth, and increase the random integer value in the original image to get more image data, the experimental results show that the proposed method is effective. (2) deep learning (Deep Learning DL) is a hot topic in machine learning, speech recognition in many fields such as image recognition, Natural Language Processing and other fields the breakthrough. Convolutional neural network (convolutional neural networks, CNN) is a kind of deep learning algorithm, based on CNN principle, training a depth CNN network model SARnet, SARnet contains two volumes Laminate, convolution kernel size is 7 x 7, two pool layer, two layer fully connected; after enhancement method combined with data set is 22 times of the original database of training, training of the network, the network model is trained by the feature extraction of the MSTAR database test set, the classification of feature extraction the SVM, the final recognition rate reached 95.68%, the recognition rate is higher than the other CNN model. (3) recently in machine learning, researchers have focused on learning some features from the unlabeled data, according to the SAR image feature extraction, the extracting regions of interest in the input image size 64 x 64; then using unsupervised K mean feature learning algorithm, combining the training database data after enhancement, learn some useful representation. By block self encoding and optimization parameters were accepted domain features of SAR image can make learning The model to learn the characteristics of variety, and proved by experiments, the mainstream recognition in unsupervised feature feature learning obtained can make SAR image recognition rate of 96.67% with the rate of K-means.

【學(xué)位授予單位】：五邑大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TN957.52

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 李君寶;楊文慧;許劍清;彭宇;;基于深度卷積網(wǎng)絡(luò)的SAR圖像目標(biāo)檢測識別[J];導(dǎo)航定位與授時;2017年01期

2 田壯壯;占榮輝;胡杰民;張軍;;基于卷積神經(jīng)網(wǎng)絡(luò)的SAR圖像目標(biāo)識別研究.[J];雷達(dá)學(xué)報;2016年03期

3 Yuan Zhan;He You;Cai Fuqing;;Bayesian edge detector for SAR imagery using discontinuity-adaptive Markov random feld modeling[J];Chinese Journal of Aeronautics;2013年06期

4 孫志軍;薛磊;許陽明;孫志勇;;基于多層編碼器的SAR目標(biāo)及陰影聯(lián)合特征提取算法[J];雷達(dá)學(xué)報;2013年02期

5 張向榮;陽春;焦李成;;基于Laplacian正則化最小二乘的半監(jiān)督SAR目標(biāo)識別[J];軟件學(xué)報;2010年04期

6 張靜;王國宏;楊智勇;劉福太;;基于二維子分類鑒別分析的SAR圖像識別方法研究[J];電子學(xué)報;2010年04期

7 趙志宏;楊紹普;馬增強(qiáng);;基于卷積神經(jīng)網(wǎng)絡(luò)LeNet-5的車牌字符識別研究[J];系統(tǒng)仿真學(xué)報;2010年03期

8 李翊;張靜;吳凌華;楊迎化;;一種基于改進(jìn)核主成分分析的SAR圖像識別方法研究[J];海軍航空工程學(xué)院學(xué)報;2009年03期

9 韓萍,吳仁彪,王兆華,王蘊(yùn)紅;基于KPCA準(zhǔn)則的SAR目標(biāo)特征提取與識別[J];電子與信息學(xué)報;2003年10期

，

本文編號：1756040