-استخراج ویژگی با کمک تجزیه به عناصر تنک به منظور طبقه بندی چهره

نوع مقاله : مقاله پژوهشی

نویسنده

استادیار، دانشکده برق و کامپیوتر، دانشگاه صنعتی همدان، همدان رایانامه: h.doosti@hut.ac.ir

چکیده

 در سـالهای اخیر، اسـتخراج ویژگی بهعنوان یک مرحله میانی در طبقهبندی موردتوجه پژوهشـگران بودهاسـت.
در این مقاله، یک روش نوین بهمنظور اســتخراج ویژگی بانظارت با کمک تجزیه به عناصــر تنک پیشــنهاد میشــود.
الگوریتم پیشـنهادی شـامل دو مرحله اسـت که در مرحله اول اطلاعات مشـترک دادهها در یک ماتریس با مرتبه کم قرار
میگیرد و در مرحله دوم یک روش اســتخراج ویژگی خطی مانند نگاشــت حفظ موقعیت مکانی بهمنظور اســتخراج
نهایی ویژگیها مورد اســتفاده قرار میگیرد. ســپس ویژگیهای اســتخراج شــده به طبقهبند ماشــینبردارپشــتیبان داده
میشـود. بهمنظور سـنجش صـحت روش پیشـنهادی، از سـه مجموعه داده اسـتفاده میشـود. نتایج آزمایش نشـاندهنده
برتری روش پیشنهادی نسبت به برخی از روشهای مدرن استخراج ویژگی است
 

کلیدواژه‌ها


عنوان مقاله [English]

Feature extraction using sparse component decomposition for face classification

نویسنده [English]

  • Hamid Reza Shahdoosti
Assistance Prof. Electrical and Computer Engineering Department,, Hamedan University of tecnology. Hamedan, Iran. Email: h.doosti@hut.ac.ir
چکیده [English]

In the recent years, the feature extraction as an intermediate step in the classification, has attracted the attention of researchers. In this paper, a new supervised feature extraction method is proposed using sparse component decomposition. The proposed algorithm has two steps.In the first step, the common information of the data matrix is extracted in a low rank matrix. In he second step, a linear feature extractor method such as local preservation projection one is used to extract the final features. Then, the extracted features are fed to the support vector machine classifier. To evaluate the accuracy rate of the proposed method, three datasets are used. The results show that the proposed method outperforms compared with some state of the art methods.

کلیدواژه‌ها [English]

  • Feature extraction
  • Face classification
  • Sparse decomposition
  • Support vector machine
Baudat. G. and Anouar. F., Generalized discriminant analysis using a kernel approach, Neural Comput., 2000, vol. 12, no. 10, pp. 2385–2404. https://doi.org/10.1016/j.trb.2017.04.003
Candès, E., Li, X., Ma, Y. and Wright, J., Robust principal component analysis?, J. ACM, 2011, vol. 58, no. 3, pp. 1–37. https://doi.org/1085/j.trb.2005.15.112
Cui. Y and Fan. L, A novel supervised dimensionality reduction algorithm: Graph-based Fisher analysis, Pattern Recognition, 2012a, Volume 45, Issue 4, pp. 1471–1481. https://doi.org/1075/j.trb.2003.4.125
Cui. Y and Fan. L., Feature extraction using fuzzy maximum margin criterion, Neurocomputing, 2012b, vol. 86, no.1, pp. 52-58.  https://doi.org/1077/j.trb.2011.38.123
Elad. M and Aharon. M, Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries, IEEE Transactions on Image Processing, 2006, vol. 15,  no. 12, pp. 3736-3745. https://doi.org/1063/j.trb.2017.36.58
He. X.F., Cai. D., Yan. S.C. and Zhang. H.J, Neighborhood preserving embedding, in: Proceedings of the Tenth IEEE International Conference on Computer Vision2 (ICCV 2005), 2005, pp. 1208–1213. https://doi.org/1086/j.trb.2022.16.41
Imani. M and Ghassemian. H, Feature Extraction Using Weighted Training Samples, IEEE Geoscience and Remote Sensing Letters, 2015, vol,.12, no. 7, pp.1387-1386. https://doi.org/1052/j.trb.2010.23.130
Jolliffe. I.T., Principal Component Analysis, second ed., Springer-Verlag, New York, 2002. https://doi.org/1075/j.trb.2014.30.74
Mika. S., Kernel Fisher Discriminant, PhD Thesis, University of Technology, Berlin, 2002. https://doi.org/1029/j.trb.2018.21.111
Roweis. S.T. and Saul. L.K., Nonlinear dimensionality reduction by locally linear embedding, Science, 2000, vol. 290 no. 5500,  pp. 2323–2326. https://doi.org/1085/j.trb.2015.22.36
Song. B, Li. J, Mura. M.D, Li. P, Plaza. A, Bioucas-Dias. J.M, Benediktsson. J.A,  and Chanussot. J., Remotely Sensed Image Classification Using Sparse Representations of Morphological Attribute Profiles, IEEE Transactions on Geoscience and Remote Sensing 2014, vol. 52 ,  no. 8, pp. 5122-5136. https://doi.org/1052/j.trb.2002.30.122
Wan. T, Zhu. C and Qin. Z, Multifocus image fusion based on robust principal component analysis, Pattern Recognition Letters, 2013, vol.34. no. 9, pp. 1001-1008. https://doi.org/1024/j.trb.2017.22.111
Wright, J., Ganesh, A., Rao, S. and Ma, Y., Robust principal component analysis: Exact recovery of corrupted low-rank matrices via convex optimization. In: The Proc. of the Conf. on Neural Information Processing Systems, 2009, pp. 1–9. https://doi.org/1012/j.trb.2023.31.123
Yang. J., Frangi. F., Yang. J.Y., Zhang. D. and Jin. Z., KPCA plus LDA: a complete kernel Fisher discriminant framework for feature extraction and recognition, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, vol. 27, no. 2, pp. 230–244. https://doi.org/1062/j.trb.2011.30.69
Zhu. X.X. and Bamler. R, A sparse image fusion algorithm with application to pan-sharpening, IEEE Transactions on Geoscience and Remote Sensing., 2013, vol. 51, no. 5, pp. 2827–2836. https://doi.org/1077/j.trb.2004.22.19
CAPTCHA Image