利用穩(wěn)健字典學(xué)習(xí)的雷達(dá)高分辨距離像目標(biāo)識別算法
doi: 10.11999/JEIT141227
基金項目:
國家自然科學(xué)基金(61372132, 61201292),新世紀(jì)優(yōu)秀人才支持計劃(NCET-13-0945),重點實驗室基金和中央高?;究蒲袠I(yè)務(wù)費專項資金資助課題
Radar High Resolution Range Profile Target Recognition Algorithm via Stable Dictionary Learning
-
摘要: 基于字典學(xué)習(xí)算法的信號稀疏表示被廣泛應(yīng)用于信號處理領(lǐng)域。由于字典原子間存在冗余性,求解信號的稀疏表示會受到觀測信號中擾動分量的影響,從而帶來表示的不確定性,不利于雷達(dá)高分辨距離像(HRRP)目標(biāo)識別任務(wù)。針對這一問題,該文提出一種穩(wěn)健字典學(xué)習(xí)(SDL)算法,通過邊緣化信號丟失,構(gòu)建穩(wěn)健損失函數(shù)用于學(xué)習(xí)自適應(yīng)字典。該算法利用距離像在散射點不發(fā)生越距離單元走動的方位幀內(nèi)具有結(jié)構(gòu)相似性,約束臨近訓(xùn)練樣本間稀疏表示的非零元素位置相同,并通過結(jié)構(gòu)化稀疏約束選擇最優(yōu)子字典用于測試樣本的分類?;趯崪yHRRP數(shù)據(jù)的實驗結(jié)果驗證了所提算法的有效性。
-
關(guān)鍵詞:
- 雷達(dá)自動目標(biāo)識別 /
- 高分辨距離像 /
- 穩(wěn)健字典學(xué)習(xí) /
- 邊緣化信號丟失 /
- 穩(wěn)健稀疏表示
Abstract: The sparse representation of signal via dictionary learning algorithms is widely used in signal processing field. Since there is redundancy in the new space defined by overcomplete dictionary atoms, the problem of finding sparse representations may bring the uncertainty and ambiguity in the presence of unknown amplitude perturbations, which is unfavorable to radar High Resolution Range Profile (HRRP) target recognition task. To deal with this issue, this paper proposes a novel algorithm called Stable Dictionary Learning (SDL), which constructs a robust loss function via marginalizing dropout to learn a stable adaptive dictionary. The algorithm considers the structure similarity among the adjacent HRRPs without scatterers motion through range cells, and enforces the constraints that the sparse representations of adjacent HRRPs should have the same supports. Moreover, SDL utilizes the structured sparse regularization learned in the training phase to automatically select the optimal sub-dictionary basis vectors, which is used for the classification of the test sample. Experimental results on measured radar HRRP dataset validate the effectiveness of the proposed method. -
Chen B, Liu H W, Chai J, et al.. Large margin feature weighting method via linear programming[J]. IEEE Transactions on Knowledge Data Engineering, 2009, 21(10): 1475-1488. 潘勉, 王鵬輝, 杜蘭, 等. 基于TSB-HMM模型的雷達(dá)高分辨距離像目標(biāo)識別算法[J]. 電子與信息學(xué)報, 2013, 35(7): 1547-1554. Pan Mian, Wang Peng-hui, Du Lan, et al.. Radar HRRP target recognition based on truncated stick-breaking hidden Markov model[J]. Journal of Electronics Information Technology, 2013, 35(7): 1547-1554. Chai J, Liu H W, and Bao Z. Combinatorial discriminant analysis: supervised feature extraction that integrates global and local criteria[J]. Electronics Letters, 2009, 45(18): 934-935. Du L, Liu H W, Bao Z, et al.. Radar HRRP target recognition based on higher-order spectra[J]. IEEE Transactions on Signal Processing, 2005, 53(7): 2359-2368. Du L, Liu H W, Bao Z, et al.. Radar automatic target recognition using complex high-resolution range profiles[J]. IET Radar, Sonar, Navigation, 2007, 1(1): 18-26. Du L, Liu H W, Wang P H, et al.. Noise robust radar HRRP target recognition based on multitask factor analysis with small training data size[J]. IEEE Transactions on Signal Processing, 2012, 60(7): 3546-3559. 馮博, 杜蘭, 張學(xué)峰, 等. 基于字典學(xué)習(xí)的雷達(dá)高分辨距離像目標(biāo)識別[J]. 電波科學(xué)學(xué)報, 2012, 27(5): 897-905. Feng Bo, Du Lan, Zhang Xue-feng, et al.. Radar HRRP target recognition based on dictionary learning[J]. Chinese Journal of Radio Science, 2012, 27(5): 897-905. To?ic? I and Frossard P. Dictionary learning[J]. IEEE Signal Processing Magazine, 2011, 28(2): 2738. Donoho D L, Elad M, and Temlyakov V. Stable recovery of sparse overcomplete representations in the presence of noise[J]. IEEE Transactions on Information Theory, 2006, 52(1): 6-18. Aharon M, Elad M, and Bruckstein A. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322. Wang H, Nie F P, and Huang H. Robust and discriminative self-taught learning[C]. International Conference on Machine Learning (ICML-13), Atlanta, Georgia, USA, 2013: 298-306. Bengio Y. Neural Networks: Tricks of the Trade[M]. Berlin Heidelberg: Springer, 2012: 437-478. Wan L, Zeiler M, Zhang S X, et al.. Regularization of neural networks using DropConnect[J]. JMLR WCP, 2013, 28(3): 1058-1066. Srivastava N. Improving neural networks with dropout[D]. [Ph.D. dissertation], University of Toronto, 2013. -
計量
- 文章訪問數(shù): 1564
- HTML全文瀏覽量: 176
- PDF下載量: 522
- 被引次數(shù): 0