1 / 54

CS479/679 Pattern Recognition Dr. George Bebis

Feature Selection Jain, A.K.; Duin , P.W.; Jianchang Mao, “Statistical pattern recognition: a review”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 1, pp. 4-37, Jan. 2000. Section 4.2 Genetic Algorithms Chapter 7 ( Duda et al.) – Section 7.5.

kaya
Download Presentation

CS479/679 Pattern Recognition Dr. George Bebis

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Feature SelectionJain, A.K.; Duin, P.W.; Jianchang Mao, “Statistical pattern recognition: a review”, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 1, pp. 4-37, Jan. 2000.Section 4.2 Genetic AlgorithmsChapter 7 (Duda et al.) – Section 7.5 CS479/679 Pattern RecognitionDr. George Bebis

  2. Feature Selection • Given a set of n features, the role of feature selection is to select a subset of size d (d<n) that leads to the smallest classification error. • Fundamentally different from dimensionality reduction (e.g.,PCA or LDA) dimensionality reduction

  3. Why is feature selection important? • Features may be expensive to obtain • You evaluate a large number of features (sensors) in the test bed and select only a few for the final implementation • You may want to extract meaningful rules from your classifier • When you transform or project, the measurement units (length, weight, etc.) of your features are lost • Features may not be numeric • A typical situation in the machine learning domain

  4. Feature Selection Methods • Feature selection is an optimization problem. • Search the space of possible feature subsets. • Pick the subset that is optimal or near-optimal with respect to an objective function.

  5. Feature Selection Methods • Feature selection is an optimization problem. • Search the space of possible feature subsets. • Pick the subset that is optimal or near-optimal with respect to a certain criterion. Search strategies Evaluationstrategies • Optimum - Filter methods • Heuristic - Wrapper methods • Randomized

  6. Search Strategies • Assuming m features, an exhaustive search would require: • Examining all possible subsets of size d. • Selecting the subset that performs the best according to the criterion function. • The number of subsets grows combinatorially, making exhaustive search impractical. • Iterative procedures are often used based on heuristics but they cannot guarantee the selection of the optimal subset.

  7. Evaluation Strategies • Filter Methods • Evaluation is independent of the classification algorithm. • The objective function evaluates feature subsets by their information content, typically interclass distance, statistical dependence or information-theoretic measures.

  8. Evaluation Strategies • Wrapper Methods • Evaluation uses criteria related to the classification algorithm. • The objective function is a pattern classifier, which evaluates feature subsets by their predictive accuracy (recognition rate on test data) by statistical resampling or cross-validation.

  9. Filter vs Wrapper Approaches

  10. Filter vs Wrapper Approaches (cont’d)

  11. Naïve Search • Sort the given n features in order of their probability of correct recognition. • Select the top d features from this sorted list. • Disadvantage • Feature correlation is not considered. • Best pair of features may not even contain the best individual feature.

  12. Sequential forward selection (SFS)(heuristic search) • First, the best single feature is selected (i.e., using some criterion function). • Then, pairs of features are formed using one of the remaining features and this best feature, and the best pair is selected. • Next, triplets of features are formed using one of the remaining features and these two best features, and the best triplet is selected. • This procedure continues until a predefined number of features are selected. SFS performs best when the optimal subset is small.

  13. Example features added at each iteration Results of sequential forward feature selection for classification of a satellite image using 28 features. x-axis shows the classification accuracy (%) and y-axis shows the features added at each iteration (the first iteration is at the bottom). The highest accuracy value is shown with a star.

  14. Sequential backward selection (SBS)(heuristic search) • First, the criterion function is computed for all n features. • Then, each feature is deleted one at a time, the criterion function is computed for all subsets with n-1 features, and the worst feature is discarded. • Next, each feature among the remaining n-1 is deleted one at a time, and the worst feature is discarded to form a subset with n-2 features. • This procedure continues until a predefined number of features are left. SBS performs best when the optimal subset is large.

  15. Example features removed at each iteration Results of sequential backwardfeature selection for classification of a satellite image using 28 features. x-axis shows the classification accuracy (%) and y-axis shows the features removed at each iteration (the first iteration is at the top). The highest accuracy value is shown with a star.

  16. Bidirectional Search (BDS) • BDS applies SFS and SBS simultaneously: • SFS is performed from the empty set • SBS is performed from the full set • To guarantee that SFS and SBS converge to the same solution • Features already selected by SFS are not removed by SBS • Features already removed by SBS are not selected by SFS

  17. “Plus-L, minus-R” selection (LRS) • A generalization of SFS and SBS • If L>R, LRS starts from the empty set and: • Repeatedly add L features • Repeatedly remove R features • If L<R, LRS starts from the full set and: • Repeatedly removes R features • Repeatedly add L features • LRS attempts to compensate for the weaknesses of SFS and SBS with some backtracking capabilities.

  18. Sequential floating selection (SFFS and SFBS) • An extension to LRS with flexible backtracking capabilities • Rather than fixing the values of L and R, floating methods determine these values from the data. • The dimensionality of the subset during the search can be thought to be “floating” up and down • There are two floating methods: • Sequential floating forward selection (SFFS) • Sequential floating backward selection (SFBS) P. Pudil, J. Novovicova, J. Kittler, Floating search methods in feature selection, Pattern Recognition Lett. 15 (1994) 1119–1125.

  19. Sequential floating selection (SFFS and SFBS) • SFFS • Sequential floating forward selection (SFFS) starts from the empty set. • After each forward step, SFFS performs backward steps as long as the objective function increases. • SFBS • Sequential floating backward selection (SFBS) starts from the full set. • After each backward step, SFBS performs forward steps as long as the objective function increases.

  20. Classifier Pre- Processing Feature Extraction Feature Subset Feature Subset Feature Selection (GA) Feature Selection using Genetic Algorithms (GAs)(randomized search) GAs provide a simple, general, and powerful framework for feature selection.

  21. Genetic Algorithms (GAs) • What is a GA? • An optimization technique for searching very large spaces. • Inspired by the biological mechanisms of natural selection and reproduction. • Main characteristics of GAs • Global optimization technique. • Searches probabilistically using a population of structures, each properly encoded as a string of symbols. • Uses objective function information, not derivatives. CS 776 Evolutionary Computing

  22. Encoding • Encoding scheme • Transforms solutions in parameter space into finite length strings (chromosomes) over some finite set of symbols.

  23. Fitness Evaluation • Fitness function • Evaluates the goodness of a solution.

  24. GA Search Process • 10010110… 10010110… • 01100010… 01100010… • 10100100... 10100100… • 10010010… 01111001… • 01111101… 10011101… Selection Crossover Mutation Current Generation Next Generation

  25. Selection Operator • Probabilistically filters out solutions that perform poorly, choosing high performance solutions to exploit. fitness

  26. Crossover and Mutation Operators • Explore new solutions. • Crossover: information exchange between points. • Mutation:restore lost genetic material. mutated bit

  27. Genetic Feature Subset Selection • Binary encoding • Fitness evaluation Feature#1 Feature#N fitness=104accuracy + 0.4  #zeros accuracy from validation set number of features

  28. Case Study 1: Gender Classification Z. Sun, G. Bebis, X. Yuan, and S. Louis, "Genetic Feature Subset Selection for Gender Classification: A Comparison Study", IEEE Workshop on Applications of Computer Vision, pp. 165-170, Orlando, December 2002. • Determine the gender of a subject from facial images. • Race, age, facial expression, hair style, etc.

  29. Feature Extraction Using PCA • PCA maps the data in a lower-dimensional space using a linear transformation. • The columns of the projection matrix are the “best” eigenvectors (i.e., eigenfaces) of the covariance matrix of the data.

  30. EV#1 EV#2 EV#3 EV#4 EV#5 EV#6 EV#8 EV#10 EV#12 EV#14 EV#19 EV#20 Which eigenvectors encode mostly gender information? IDEA: Use GAs to search the space of eigenvectors!

  31. Dataset • 400 frontal images from 400 different people • 200 male, 200 female • Different races, lighting conditions, and facial expressions • Images were registered and normalized • No hair information • Normalized to account for different lighting conditions

  32. Experiments • Classifiers • LDA • Bayes classifier • Neural Networks (NNs) • Support Vector Machines (SVMs) • Three-fold cross validation • Training set: 75% of the data • Validation set: 12.5% of the data • Test set: 12.5% of the data • Comparison with SFBS

  33. Error Rates 22.4% 17.7% 14.2% 13.3% 11.3% 9% 8.9% 6.7% 4.7% ERM: error rate using top eigenvectors ERG: error rate using GA selected eigenvectors

  34. Ratio of Features - Information Kept 69.0% 61.2% 42.8% 38% 36.4% 31% 32.4% 17.6% 13.3% 8.4% RN: percentage of eigenvectors in the feature subset selected. RI: percentage of information contained in the eigenvector subset selected.

  35. Histograms of Selected Eigenvectors (b) Bayes (a) LDA (c) NN (d) SVMs

  36. Reconstructed Images Original images Using top 30 EVs Using EVs selected by NN+GA Using EVs selected by SVM+GA Reconstructed faces using GA-selected EVs do not contain information about identity but disclose strong gender information!

  37. Comparison with SFBS Original images Top 30 EVs EVs selected by SVM+GA EVs selected by SVM+SFBS

  38. Case Study 2: Vehicle Detection Z. Sun, G. Bebis, and R. Miller, "Object Detection Using Feature Subset Selection", Pattern Recognition, vol. 37, pp. 2165-2176, 2004. low light camera rear views Non-vehicles class much larger than vehicle class. Ford Motor Company

  39. Which eigenvectors encode themost important vehicle features?

  40. Experiments • Training data set (collected in Fall 2001) • 2102 images (1051 vehicles and 1051 non-vehicles) • Test data sets (collected in Summer 2001) • 231 images (vehicles and non-vehicles) • SVM for classification • Three-fold cross-validation • Comparison with SFBS

  41. Error Rate 6.49%

  42. Histograms of Selected Eigenvectors SFBS-SVM GA-SVM Number of eigenvectors selected by SBFS: 87 (43.5% information) Number of eigenvectors selected by GA: 46 (23% information)

  43. Vehicle Detection Original Top 50 EVs EVs selected by SFBS Evs selected by GAs Reconstructed images using the selected feature subsets. - Lighting differences have been disregarded by the GA approach.

  44. Case Study 3:Visible-Thermal IR Fusion for Face Recognition G. Bebis, A. Gyaourova, S. Singh, and I. Pavlidis, "Face Recognition by Fusing Thermal Infrared and Visible Imagery", Image and Vision Computing, vol. 24, no. 7, pp. 727-742, 2006.

  45. Visible vs Thermal IR • Visible spectrum • High resolution, less sensitive to the presence of eyeglasses. • Sensitive to changes in illumination and facial expression. • Thermal IR spectrum • Robust to illumination changes and facial expressions. • Low resolution, sensitive to air currents, face heat patterns, aging, and the presence of eyeglasses (i.e., glass is opaque to thermal IR).

  46. Feature Extraction Reconstruct Image Fusion Using Genetic Algorithms Fused Image How should we fuse information from visible and thermal IR? We have tested two feature extraction methods: - PCA and Wavelets

  47. Dataset Co-registered thermal infrared and visible images. Variations among images: -Eyeglasses / No eyeglasses. - Different illuminations – front, left, right. - Different facial expressions – smile, frown, surprise and vowels Equinox database

  48. Frontal Illumination Lateral illumination DATA EFnG EFG ELG ELnG EnG EG No Glasses Glasses Eyeglasses/Illumination Tests

  49. Experiments glasses no glasses test database glasses no glasses Eyeglasses Tests Illumination Tests

  50. Results

More Related