基于Fisher約束和字典對(duì)的圖像分類
doi: 10.11999/JEIT160296
基金項(xiàng)目:
國(guó)家973計(jì)劃項(xiàng)目(2014CB340400)，天津市自然科學(xué)基金(15JCYBJC15500)
Image Classification Based on Fisher Constraint and Dictionary Pair
Funds:
The National 973 Program of China (2014CB340400), The Natural Science Foundation of Tianjin (15JCYBJC15500)
-
摘要: 基于稀疏表示的分類方法由于其所具有的簡(jiǎn)單性和有效性獲得了研究者的廣泛關(guān)注，然而如何建立字典原子與類別信息間的聯(lián)系仍然是一個(gè)重要的問(wèn)題，與此同時(shí)大部分稀疏表示分類方法都需要求解受范數(shù)約束的優(yōu)化問(wèn)題，使得分類任務(wù)的計(jì)算較復(fù)雜。為解決上述問(wèn)題，該文提出一種新的基于Fisher約束的字典對(duì)學(xué)習(xí)方法。新方法聯(lián)合學(xué)習(xí)結(jié)構(gòu)化綜合字典和結(jié)構(gòu)化解析字典，然后通過(guò)樣本在解析字典上的映射直接求解稀疏系數(shù)矩陣；同時(shí)采用Fisher判別準(zhǔn)則編碼系數(shù)使系數(shù)具有一定的判別性。最后將新方法應(yīng)用到圖像分類中，實(shí)驗(yàn)結(jié)果表明新方法在提高分類準(zhǔn)確率的同時(shí)還大大降低了計(jì)算復(fù)雜度，相較于現(xiàn)有方法具有更好的性能。Abstract: Classification method based on sparse representation has won wide attention because of its simplicity and effectiveness, while how to adaptively build the relationship between dictionary atoms and class labels is still an important open question, at the same time most of the sparse representation classification methods need to solve a norm constraint optimization problem, which increases the computational complexity in the classification task. To address this issue, this paper proposes a novel Fisher constraint dictionary pair learning method to jointly learn a structured synthesis dictionary and a structured analysis dictionary, then directly obtains the sparse coefficient matrix by analysis dictionary. In this paper, the Fisher criterion is used to encode the coefficients. Finally the new method is applied to image classification task, the experimental results show that the new method not only improves the accuracy of classification but also greatly reduces the computational complexity. Compared with the existing methods, the new method has better performance.
-
Key words:
- Image classification /
- Sparse representation /
- Dictionary pair /
- Fisher constraint
-
RUBINSTEIN R, BRUCKSTEIN A, ELAD M, et al. Dictionaries for sparse representation modeling[J]. Proceedings of the IEEE, 2010, 98(6): 1045-1057. doi: 10.1109/JPROC.2010.2040551. GAO Shenghua, TSANG I, and MA Yi. Learning category- specific dictionary and shared dictionary for fine-grained image categorization[J]. IEEE Transactions on Image Processing, 2014, 23(2): 623-634. doi: 10.1109/TIP.2013. 2290593. 宋相法, 焦李成. 基于稀疏編碼和集成學(xué)習(xí)的多示例多標(biāo)記圖像分類方法[J]. 電子與信息學(xué)報(bào), 2013, 35(3): 622-626. doi: 10.3724/SP.J.1146.2012.01218. SONG Xiangfa and JIAO Licheng. A multi-instance multi-label image classification method based on sparse coding and ensemble learning[J]. Journal of Electronics Information Technology, 2013, 35(3): 622-626. doi: 10.3724/ SP.J.1146.2012.01218. AHARON M, ELAD M, and BRUCKSTEIN A. K-SVD: an algorithm for designing of overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322. doi: 10.1109/TSP. 2006.881199. MA Long, WANG Chunheng, XIAO Baihua, et al. Sparse representation for face recognition based on discriminative low-rank dictionary learning[C]. IEEE Conference on Computer Vision and Pattern Recognition, Providence, USA, 2012: 2586-2593. RAMIREZ I, SPRECHMANN P, SAPIRO G, et al. Classification and clustering via dictionary learning with structured incoherence and shared features[C]. IEEE Conference on Computer Vision and Pattern Recognition, San Francisco, CA, USA, 2010: 3501-3508. NING Zhou and FAN Jianping. Jointly learning visually correlated dictionaries for large-scale visual recognition applications[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(4): 715-730. doi: 10.1109/ TPAMI.2013.189. JIANG Zhuolin, LIN Zhe, LARRY S, et al. Label consistent K-SVD: Learning a discriminative dictionary for recognition [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(11): 2651-2664. doi: 10.1109/TPAMI. 2013.88. YANG Meng, ZHANG Lei, FENG Xiangchu, et al. Sparse representation based fisher discrimination dictionary learning for image classication[J]. International Journal of Computer Vision, 2014, 109(3): 209-232. doi: 10.1007/s11263-014- 0722- 8. 練秋生, 石保順, 陳書貞. 字典學(xué)習(xí)模型、算法及其應(yīng)用研究進(jìn)展[J]. 自動(dòng)化學(xué)報(bào), 2015, 41(2): 240-260. doi: 10.16383/ j.aas.2015.c140252. LIAN Qiusheng, SHI Baoshun, and CHEN Shuzhen. Research advances on dictionary learning models, algorithms and applications[J]. Acta Automatica Sinica, 2015, 41(2): 240-260. doi: 10.16383/j.aas.2015.c140252. RUBINSTEIN R, PELEG T, and ELAD M. Analysis K-SVD: A dictionary-learning algorithm for the analysis sparse model[J]. IEEE Transactions on Signal Processing, 2013, 61(3): 661-677. doi: 10.1109/TSP.2012.2226445. CHEN Yunjin, RANFTL R, and POCK T. Insights into analysis operator learning: From patch-based sparse models to higher order MRFs[J]. IEEE Transactions on Image Processing, 2014, 23(3): 1060-1072. doi: 10.1109/TIP.2014. 2299065. GU Shuhang, ZHANG Lei, ZUO Wangmeng, et al. Projective dictionary pair learning for pattern classification[C]. Advances in Neural Information Processing System, Vancouver, BC, Canada, 2014, 1: 793-801. RAKOTOMAMONJY A. Applying alternating direction method of multipliers for constrained dictionary learning[J]. Neurocomputing, 2013, 61(3): 126-136. doi: 10.1016/j. neucom.2012.10.024. CAI Sijia, ZUO Wangmeng, ZHANG Lei, et al. Support vector guided dictionary learning[C]. European Conference on Computer Vision, Zurich, Switzerland, 2014, 8692: 624-639. GORSKI J, PFEUFFER F, and KLAMROTH K. Biconvex sets and optimization with biconvex functions: a survey and extensions[J]. Mathematical Methods of Operations Research, 2007, 66(3): 373-407. doi: 10.1007/s00186-007-0161-1. -
計(jì)量
- 文章訪問(wèn)數(shù): 1313
- HTML全文瀏覽量: 146
- PDF下載量: 527
- 被引次數(shù): 0
下載:
