Pattern recognition
- 2. Pattern Recognition Swarnava Sen MCKV Institute of Engineering, Liluah IT/2014/028
- 3. What is pattern ? o A pattern is an abstract object, or a set of measurements describing a physical object.
- 4. What is a pattern CLass? o A pattern class (or category) is a set of patterns sharing common attributes. o A collection of “Similar” (not necessarily identical) objects. o During recognition given objects are assigned to prescribed classes.
- 5. What is pattern reCOGnitiOn ? o Theory, Algorithms, Systems to put Patterns into Category. o Relate Perceived Pattern to Previously Perceived Patterns. o Learn to distinguish patterns of interest from their background.
- 6. hUMan perCeptiOn o Humans have developed highly sophisticated skills for sensing their environment and taking actions according to what they observe, e.g., o Recognizing a face. o Understanding spoken words. o Reading handwriting. o Distinguishing fresh food from its smell. o We would like to give similar capabilities to machines.
- 7. • Handwritten: sorting letters by postal code. • Printed texts: reading machines for blind people, digitalization of text documents. Optical Character Recognition (OCR) • Face recognition, verification, retrieval. • Finger prints recognition. • Speech recognition. Biometrics • Medical diagnosis: X-Ray, EKG (ElectroCardioGraph) analysis. Diagnostic systems • Automated Target Recognition (ATR). • Image segmentation and analysis (recognition from aerial or satelite photographs). Military applications ExamplEs of applications
- 8. Grid by Grid comparison
- 9. Grid by Grid comparison A A B 0 0 1 0 0 0 1 0 0 1 1 1 1 0 0 1 1 0 0 1 0 1 1 0 0 1 1 0 0 1 1 0 1 0 0 1 1 0 0 1 No of Mismatch= 3
- 10. Grid by Grid comparison A A BGrid by Grid Comparison
- 11. Grid by Grid comparison A A B 0 0 1 0 0 0 1 0 0 1 1 1 1 0 0 1 1 0 0 1 1 1 1 0 0 1 0 1 0 1 1 1 0 1 0 1 1 1 1 0 No of Mismatch= 9
- 12. problEm with Grid by Grid comparison o Time to recognize a pattern - Proportional to the number of stored patterns (Too costly with the increase of number of patterns stored) A-Z a-z 0-9 Solution Artificial Intelligence */-+1@#
- 13. pattern recognition o Two Phase : Learning and Detection. o Time to learn is higher. • Driving a car o Difficult to learn but once learnt it becomes natural. o Can use AI learning methodologies such as: • Neural Network. • Machine Learning.
- 14. Learning How can machine learn the rule from data? Supervised learning: a teacher provides a category label or cost for each pattern in the training set. Unsupervised learning: the system forms clusters or natural groupings of the input patterns.
- 15. o Classification (known categories) o Clustering (creation of new categories) cLassification vs. cLustering Category “A” Category “B” Clustering (Unsupervised Classification) Classification (Supervised Classification)
- 16. pattern recognition process Post- processing Classification Feature Extraction Segmentation Sensing input Decision
- 17. pattern recognition process o Sensing: • Measurements of physical variables. • Important issues: bandwidth, resolution , etc. o Segmentation: • Removal of noise in data. • Isolation of patterns of interest from the background. o Feature extraction: • Finding a new representation in terms of features. o Classification • Using features and learned models to assign a pattern to a category. o Post-processing • Evaluation of confidence in decisions. Post- processing Classification Feature Extraction Segmentation Sensing input Decision
- 18. • Collecting training and testing data. Collect Data • Domain dependence. Chose Features. • Domain dependence. Chose Model • Supervised learning • Unsupervised learning. Train • Performance with future data Evaluate the Design cycLe
- 19. Demo o Online face detector demo:
- 20. Demo (Cont.) With my friend “Albert Einstein”
- 21. tHAnK YoU