1 / 17

IMAGE PRE-PROCESSING FOR CLASSIFICATION (BIOMETRIC IDENTIFICATION) BY A NEURAL NETWORK

Horizon Imaging, LLC Innovative Solutions in Image Processing. IMAGE PRE-PROCESSING FOR CLASSIFICATION (BIOMETRIC IDENTIFICATION) BY A NEURAL NETWORK Anthony Vannelli, Steve Wagner, and Ken McGarvey Horizon Imaging, LLC email: info@horizonimaging.com. Raw 512x480 Image.

belvap
Télécharger la présentation

IMAGE PRE-PROCESSING FOR CLASSIFICATION (BIOMETRIC IDENTIFICATION) BY A NEURAL NETWORK

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. Horizon Imaging, LLC Innovative Solutions in Image Processing IMAGE PRE-PROCESSING FOR CLASSIFICATION (BIOMETRIC IDENTIFICATION) BY A NEURAL NETWORK Anthony Vannelli, Steve Wagner, and Ken McGarvey Horizon Imaging, LLC email: info@horizonimaging.com

  2. Raw 512x480 Image Reduced data set Classification Output Neural Network Classifier Neural Preprocessor Horizon Imaging, LLC Innovative Solutions in Image Processing Neural Network Preprocessor and Classifier • Wavelets • PCA • Image “Zones” • Combining Networks • Feed-forward Network • Back-propagation Training • Single Hidden Layer

  3. Horizon Imaging, LLC Innovative Solutions in Image Processing Curse of dimensionality • 512x480 raw image or 245,760 inputs to network • Large neural network • Poor classification performance • Slow convergence

  4. Region of Interest Horizon Imaging, LLC Innovative Solutions in Image Processing Biometric Identification 320x160 = 51,200 pixels

  5. Horizon Imaging, LLC Innovative Solutions in Image Processing Preprocessing Techniques • Non-parametric • “Holistic” • Data-driven • No Hand Geometry • No Fidiucial Points

  6. Horizon Imaging, LLC Innovative Solutions in Image Processing Preprocessing Techniques • Principal components • Large eigen-values help to classify • Reduces dimensionality • Image Processing Zones • Divide and conquer • 2x2 zones (160x80 pixels) • 4x4 zones (80x40 pixels) • Ensemble of neural networks

  7. Horizon Imaging, LLC Innovative Solutions in Image Processing Preprocessing Techniques • Combining Neural Networks • Pick the network with the “best fit” • Average the network outputs • Voting Scheme

  8. Neural Net #1    y1      1 Neural Net #2 Input Vector   Combined Output    =  i   y2  2   N yN > T Neural Net #N     yN     Horizon Imaging, LLC Innovative Solutions in Image Processing Figure 3. Voting scheme to combine Neural Networks y1 > T y2 > T 0 for yi T  i = 1 for yi > T Voting Scheme to Combine Networks

  9. Horizon Imaging, LLC Innovative Solutions in Image Processing Preprocessing Technique using Wavelets • Coiflet wavelet • Daubechies wavelet • Haar wavelet (averages adjacent pixels) Second-level wavelet approximation

  10. Low 2 LL Low 2 High 2 LH Image f(x,y) Low 2 HL High 2 High 2 HH Horizontal filter Vertical filter Horizon Imaging, LLC Innovative Solutions in Image Processing One-Level of a Wavelet Transform

  11. LL HL LH HH Horizon Imaging, LLC Innovative Solutions in Image Processing Third-level Wavelet Decomposition

  12. 320x160 pixels 512 x 480 Image Neural Classifier Image Preparation Wavelet Transform PCA Output Figure7. Test case with single classifier Horizon Imaging, LLC Innovative Solutions in Image Processing Test Case with Single Classifier

  13. Image 1 Neural Classifier 512 x 480 Image Output Image Preparation Wavelet Transform Combine Networks 320 x 160 pixels Image N Neural Classifier Figure8. Test case with multiple classifiers Horizon Imaging, LLC Innovative Solutions in Image Processing Test Case with Multiple Classifiers

  14. Horizon Imaging, LLC Innovative Solutions in Image Processing Test Cases • Coiflet 6-coefficient wavelet to 3 levels; 3rd level approximation image (40x20 pixels) and 3 sidebands form input to 4 neural networks with 800 inputs each. • Daubechies 6-coefficient wavelet to 3 levels; 3rd level approximation image (40x20) and 3 sidebands form input to 4 neural networks with 800 inputs each. • Coiflet 6-coefficient wavelet to 2 levels (80x40 pixels); 4 image zones fed to 4 separate neural networks with 800 inputs each.

  15. Horizon Imaging, LLC Innovative Solutions in Image Processing Test Cases • Daubechies 6-coefficient wavelet to 2 levels (80x40 pixels); 4 image zones fed to 4 separate neural networks with 800 inputs each. • Harr wavelet to 2 levels (80x40 pixels); 4 image zones fed to 4 separate neural networks with 800 inputs each. • Harr wavelet to 2 levels (80x40 pixels) and then PCA transform fed to a neural network with 512 inputs.

  16. Horizon Imaging, LLC Innovative Solutions in Image Processing Test Cases • Harr wavelet to 3 levels (40x20 pixels) fed to a neural network with 800 inputs. • Coiflet 6-coefficient wavelet to 1 level (160X80 = 12800 pixels). The first level approximation image is divided into 16 image zones (40x20 pixels per zone). The zones are fed into separate neural networks with 800 inputs each.

  17. Summary of Performance Horizon Imaging, LLC Innovative Solutions in Image Processing

More Related