【摘要】：合成孔徑雷達(dá)(Synthetic Aperture Radar,SAR)是一種全天時(shí)、全天候、主動(dòng)對(duì)地觀測(cè)傳感器,實(shí)現(xiàn)SAR圖像目標(biāo)識(shí)別具有重要意義。由于SAR圖像獲取成本高以及SAR圖像目標(biāo)姿態(tài)敏感性,導(dǎo)致用于目標(biāo)識(shí)別的帶標(biāo)簽SAR圖像樣本不完備,對(duì)SAR圖像目標(biāo)識(shí)別帶來(lái)挑戰(zhàn)。多任務(wù)學(xué)習(xí)(Multi-task Learning,MTL)利用不同信息源或特征,同時(shí)學(xué)習(xí)多個(gè)回歸模型優(yōu)化參數(shù),實(shí)現(xiàn)多特征信息融合,有利于提高識(shí)別性能。本文基于SAR圖像多尺度特征的稀疏表示,研究MTL架構(gòu)中的特征選擇策略、稀疏表示、稀疏求解等問(wèn)題,主要完成工作如下:(1)為滿(mǎn)足MTL對(duì)多尺度特征在稀疏域中空間分布相似性的要求,提出一種基于稀疏向量分布相似度的特征選擇方法。首先,對(duì)驗(yàn)證集樣本進(jìn)行多尺度特征稀疏表示,在不同尺度下,按類(lèi)別統(tǒng)計(jì)稀疏度分布,定義尺度間的稀疏度分布相似度矩陣,求得對(duì)應(yīng)的相關(guān)信息熵。最后,選擇相關(guān)信息熵最大的特征子集。通過(guò)實(shí)驗(yàn)分析特征的冗余性和目標(biāo)識(shí)別率,驗(yàn)證了特征選擇方法的有效性。(2)針對(duì)訓(xùn)練樣本量不充足時(shí),稀疏表示自由度偏高,提出一種多尺度特征的局部線性約束稀疏字典優(yōu)化方法�；贛TL的架構(gòu),建立多尺度特征局部線性約束,降低稀疏表示自由度,實(shí)現(xiàn)稀疏字典的優(yōu)化,提高了樣本不充足下的目標(biāo)識(shí)別率。實(shí)驗(yàn)表明,在訓(xùn)練樣本不充足時(shí),與聯(lián)合稀疏表示相比,本文方法提升了目標(biāo)識(shí)別效果。(3)設(shè)計(jì)了一種多尺度鄰域加權(quán)的匹配追蹤算法。在MTL的架構(gòu)下,通過(guò)對(duì)殘差的多尺度稀疏向量進(jìn)行鄰域加權(quán),選擇原子,實(shí)現(xiàn)匹配追蹤,得到多尺度聯(lián)合稀疏系數(shù)。在不同尺度下按類(lèi)別稀疏重構(gòu),依據(jù)多尺度累加重構(gòu)偏差,實(shí)現(xiàn)目標(biāo)分類(lèi)。實(shí)驗(yàn)結(jié)果表明該算法的重構(gòu)精度與凸優(yōu)化方法相當(dāng)并且耗時(shí)較短。
[Abstract]:Synthetic Aperture Radar (Synthetic Aperture Radar,SAR) is a kind of all-day, all-weather, active earth observation sensor to realize SAR image target recognition. Because of the high cost of SAR image acquisition and the sensitivity of target attitude in SAR image, the labeled SAR image samples for target recognition are not complete, which brings challenges to target recognition in SAR image. Multi-task learning (Multi-task Learning,MTL) utilizes different information sources or features and simultaneously learns multiple regression models to optimize parameters so as to achieve multi-feature information fusion which can improve the recognition performance. Based on the sparse representation of multi-scale features of SAR images, this paper studies the feature selection strategy, sparse representation and sparse solution in MTL architecture. The main contributions are as follows: (1) in order to meet the requirements of MTL for spatial similarity of multi-scale features in sparse domain, a feature selection method based on sparse vector distribution similarity is proposed. Firstly, the multi-scale feature sparse representation of the validation set samples is performed. According to the different scales, the sparse degree distribution is calculated according to the category, and the similarity matrix of the sparse degree distribution between scales is defined, and the corresponding information entropy is obtained. Finally, the feature subset with the largest entropy is selected. The effectiveness of the feature selection method is verified by analyzing the redundancy of the feature and the target recognition rate. (2) the sparse representation degree of freedom is high when the training sample is not sufficient. A local linear constrained sparse dictionary optimization method with multi-scale features is proposed. Based on the framework of MTL, the local linear constraints of multi-scale features are established, the degree of freedom of sparse representation is reduced, the sparse dictionary is optimized, and the target recognition rate under insufficient samples is improved. Experiments show that the proposed method improves the target recognition performance compared with the joint sparse representation when the training samples are not sufficient. (3) A multi-scale neighborhood weighted matching tracking algorithm is designed. In the framework of MTL, the multiscale joint sparse coefficients are obtained by neighborhood weighting of the multi-scale sparse vectors of the residuals and the selection of atoms to achieve matching tracing. According to the multi-scale cumulative reconstruction deviation, the target classification can be realized by sparse reconstruction according to different scales. The experimental results show that the reconstruction accuracy of the algorithm is quite similar to that of the convex optimization method and the time is short.
【學(xué)位授予單位】：南京航空航天大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TN957.52

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 丁軍;劉宏偉;陳渤;馮博;王英華;;相似性約束的深度置信網(wǎng)絡(luò)在SAR圖像目標(biāo)識(shí)別的應(yīng)用[J];電子與信息學(xué)報(bào);2016年01期

2 徐丹蕾;杜蘭;王鵬輝;劉宏偉;;采用多任務(wù)稀疏學(xué)習(xí)的雷達(dá)HRRP小樣本目標(biāo)識(shí)別[J];西安電子科技大學(xué)學(xué)報(bào);2016年02期

3 莊福振;羅平;何清;史忠植;;遷移學(xué)習(xí)研究進(jìn)展[J];軟件學(xué)報(bào);2015年01期

4 楊文,孫洪,曹永鋒;合成孔徑雷達(dá)圖像目標(biāo)識(shí)別問(wèn)題研究[J];航天返回與遙感;2004年01期

5 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(shí)別[J];電子與信息學(xué)報(bào);2003年10期

6 吳一戎,朱敏慧;合成孔徑雷達(dá)技術(shù)的發(fā)展現(xiàn)狀與趨勢(shì)[J];遙感技術(shù)與應(yīng)用;2000年02期

相關(guān)博士學(xué)位論文 前2條

1 崔宗勇;合成孔徑雷達(dá)目標(biāo)識(shí)別理論與關(guān)鍵技術(shù)研究[D];電子科技大學(xué);2015年

2 李飛;雷達(dá)圖像目標(biāo)特征提取方法研究[D];西安電子科技大學(xué);2014年

，

本文編號(hào)：2280501