DOI QR코드

DOI QR Code

PCA 기반 특징 되먹임을 이용한 중요 영역 추출

Extraction of Important Areas Using Feature Feedback Based on PCA

  • 투고 : 2020.10.10
  • 심사 : 2020.11.03
  • 발행 : 2020.12.30

초록

본 논문에서는 손글씨 숫자 데이터셋, 얼굴 데이터셋의 중요영역 추출을 위한 PCA 기반의 특징되먹임방법을 제안한다. 이전의 LDA 기반의 특징되먹임 방법을 확장하여 PCA 기반 특징되먹임 방법이 제안된다. 제안된 방법에서 데이터에 차원 축소 머신러닝 알고리듬 중 하나인 PCA 기법을 적용하여 데이터를 중요한 특징 차원들로 축소한다. 차원 축소과정에서 도출되는 weight를 통해 축소된 각 차원 축에서의 데이터 중요 지점을 확인한다. 각 차원 축은 축의 고유값의 크기에 따라 전체 데이터에서의 가중치가 다르다. 이에 각 차원 축의 고유값의 크기에 비례하는 가중치를 부여하여 각 차원 축에서의 데이터 중요 지점을 합하는 연산 과정을 거친다. 연산 과정을 통해 얻어진 데이터에 Threshold를 적용하여 데이터의 중요 영역을 구한다. 그 후 도출된 데이터의 중요 영역에 원본데이터로 역매핑을 유도하여 원본 데이터 공간에서 중요영역을 선택한다. MNIST 데이터셋에 대한 실험 결과를 확인하고 기존의 LDA 기반의 특징되먹임 방법을 통한 결과와 비교를 하여 PCA기반 특징되먹임을 기반한 패턴 인식 방법의 유효성과 가능성을 확인한다.

In this paper, we propose a PCA-based feature feedback method for extracting important areas of handwritten numeric data sets and face data sets. A PCA-based feature feedback method is proposed by extending the previous LDA-based feature feedback method. In the proposed method, the data is reduced to important feature dimensions by applying the PCA technique, one of the dimension reduction machine learning algorithms. Through the weights derived during the dimensional reduction process, the important points of data in each reduced dimensional axis are identified. Each dimension axis has a different weight in the total data according to the size of the eigenvalue of the axis. Accordingly, a weight proportional to the size of the eigenvalues of each dimension axis is given, and an operation process is performed to add important points of data in each dimension axis. The critical area of the data is calculated by applying a threshold to the data obtained through the calculation process. After that, induces reverse mapping to the original data in the important area of the derived data, and selects the important area in the original data space. The results of the experiment on the MNIST dataset are checked, and the effectiveness and possibility of the pattern recognition method based on PCA-based feature feedback are verified by comparing the results with the existing LDA-based feature feedback method.

키워드

참고문헌

  1. M. Turk and A. Pentland, "Eigenfaces for recognition," J. Cognitive Neurosci. vol. 3, no. 1, pp. 71-86, 1991 https://doi.org/10.1162/jocn.1991.3.1.71
  2. W. Lu, KN. Plataniotis, and N. Venetsanopoulos, "Face recognition using LDA -based algorithm," IEEE Trans. Neural Network, vol. 14, no. 1, pp. 195-200, 2003. https://doi.org/10.1109/TNN.2002.806647
  3. T. V. Bandos, L. Bruzzone and G. Camps-Valls, "Classification of hyperspectral images with regularized linear discriminant analysis", IEEE Trans. Geosci. Remote Sensing, vol. 47, no. 3, pp. 862-873, 2009. https://doi.org/10.1109/TGRS.2008.2005729
  4. G.-M. Jeong, H.-S. Ahn, S.-I. Choi, N. Kwak, C. Moon Pattern recognition using feature feedback: application to face recognition International Journal of Control, Automation and Systems, 8, pp. 141-148, 2010 https://doi.org/10.1007/s12555-010-0118-7
  5. Su-Hyun Kim, Sang-Il Choi, Sung-Han Bae, Young-Dae Lee, Gu-Min Jeong, "Pattern Recognition using Feature Feedback : Performance Evaluation for Feature Mask" The Institute of Internet, Broadcasting and Communication Vol 10.5, pp 179-185, October 2010
  6. Sang-Il Choi, Su-Hyun Kim, Yoonseok Yang, Gu-Min Jeong, "Data Refinement and Channel Selection for a Portable E-Nose System by the Use of Feature Feedback" Sensors 10387-10400 October 2010 https://doi.org/10.3390/s101110387
  7. Alok Sharma, Kuldip K. Paliwal "Linear discriminant analysis for the small sample size problem: an overview", International Journal of Machine Learning and Cybernetics, 6, pp 443-454, 2015 https://doi.org/10.1007/s13042-013-0226-9
  8. Sebastian Bach, Alexander Binder, Gregoire Montavon,Frederick Klauschen,Klaus-Robert Muller, Wojciech Samek "On pixel-wise explanations for non-linear classfier decisions by layerwise relevance propagation" PloS One e0130140 July 2015 https://doi.org/10.1371/journal.pone.0130140
  9. Selvaraju R.R, CogsWell M, Das A, Vedantam R, Parikh D, Batra D "Grad-CAM : Visual Explanations from Deep Networks via Gradient-based Localization" Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp 618-626 2017
  10. Y. LeCun, "The MNIST database of handwrittendigits," http://yann.lecun.com/exdb/mnist