2.2 Whole Images
About1. Introduction2. Overview3. GUI4. Image Signatures5. Unsupervised Filters6. Results & Analysis7. BioFilters8. NeuralFilters9. Duplicated Documents10. Face Recognition11. Auto Part Recognition12. Dynamic Library13. NeuralNet Filter14. Segment Variation15. TV Advertisements16. Counting & Tracking17. Image PreProcessing18. Image Processing19. Batch Job20. Parameters21. Input Option22. Application Developers23. Reference Manual24. Support Services25. Readme.txt

2.1 Internal Structures 
2.2 Whole Images 
2.3 Beyond Whole Images 
2.4 Advanced Users 

 

[Home][2. Overview][2.2 Whole Images]

 

2.2   Chapter Overview – Whole Images

Chapter 3 introduces the ImageFinder’s Graphical User Interface (GUI). This chapter introduces 1:1 Matching, 1:N Matching, and N:N Matching, and shows you how to enter data into the ImageFinder.

Chapter 4 introduces Image Signatures. Features of an image are computed and the collection of features is grouped into a signature.

Chapter 5 introduces Unsupervised Matching. Unsupervised Matching matches two whole images. The Unsupervised Matching is based on the image signatures alone. The Unsupervised Matching is used to assess a problem.

  •    In general, for an easy problem, Unsupervised Matching will do well.
  •    For a hard problem, Unsupervised Matching will NOT do well.
  •    The advantage of the Unsupervised Matching is that it is easy to use.
  •    The disadvantage of the Unsupervised Matching is that it has a low identification rate.

Chapter 6 discusses Result Analysis & Displays.

Chapter 7 introduces the BioFilter. BioFilter matches two whole images. BioFilter is better than Unsupervised Matching, but it requires a process called training. Training teaches the BioFilter who should match with whom. The BioFilter learns how to match the image features.

  •    The advantage of the BioFilter is that it does not require a lot of training data.
  •    The disadvantage of the BioFilter is that it has a lower identification rate than the Neural Filter.

Chapter 8 introduces the Neural Filter. The Neural Filter matches two whole images. The Neural Filter is better than BioFilter, but it requires far more training data. The Neural Filter learns how to match the image features from a huge amount of data.

  •    The advantage of the Neural Filter is that it is accurate.
  •    The disadvantage of the Neural Filter is that it requires a large volume of training data.

Chapter 9 presents an example: finding duplicated Document images.

Chapter 10 presents an example: identifying Faces from a photo ID.

Chapter 11 presents an example: identifying Auto Parts.

Chapter 12, Dynamic Library, briefly describes the matching process where the Master Library is constantly updated via insertion and deletion.

 

[Home][About][1. Introduction][2. Overview][3. GUI][4. Image Signatures][5. Unsupervised Filters][6. Results & Analysis][7. BioFilters][8. NeuralFilters][9. Duplicated Documents][10. Face Recognition][11. Auto Part Recognition][12. Dynamic Library][13. NeuralNet Filter][14. Segment Variation][15. TV Advertisements][16. Counting & Tracking][17. Image PreProcessing][18. Image Processing][19. Batch Job][20. Parameters][21. Input Option][22. Application Developers][23. Reference Manual][24. Support Services][25. Readme.txt]

Copyright (c) 1998 - 2006 Attrasoft, Inc. All rights reserved.

gina@attrasoft.com