CSE  Sargur Srihari  Introduction to Pattern Recognition Sargur N
165K - views

CSE Sargur Srihari Introduction to Pattern Recognition Sargur N

Srihari sriharicedarbuffaloedu Dept of Computer Science Engineering State University of New York at Buffalo brPage 2br CSE 555 Sargur Srihari 2 What is a Pattern A pattern is the opposite of chaos it is an entity vaguely defined that could be given

Download Pdf

CSE Sargur Srihari Introduction to Pattern Recognition Sargur N

Download Pdf - The PPT/PDF document "CSE Sargur Srihari Introduction to Pat..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.

Presentation on theme: "CSE Sargur Srihari Introduction to Pattern Recognition Sargur N"— Presentation transcript:

Page 1
CSE 555: Sargur Srihari 1 Introduction to Pattern Recognition Sargur N. Srihari srihari@cedar.buffalo.edu Dept. of Computer Science & Engineering State University of New York at Buffalo
Page 2
CSE 555: Sargur Srihari 2 What is a Pattern? A pattern is the opposite of chaos; it is an entity vaguely defined, that could be given a name. A pattern is an abstract object, such as a set of measurements describing a physical object.
Page 3
CSE 555: Sargur Srihari 3 Examples of Patterns Handwritten Characters Postnet ar Code Fingerprint UPC BarCode Animal Footprint

Data Trend
Page 4
CSE 555: Sargur Srihari 4 PR Definitions heory, Algorithms, Systems to put Patterns into Categories lassification of Noisy or Complex Data elate Perceived Pattern to Previously Perceived Patterns
Page 5
CSE 555: Sargur Srihari 5 Example Problem: Handwritten Digit Recognition andcrafted rules will result in large no of rules and exceptions etter to have a machine that learns from a large training set Wide variability of same numeral
Page 6
CSE 555: Sargur Srihari 6 Role of Machine Learning rincipled way of building high performance information

processing systems M v L has origins in Computer Science R has origins in Engineering hey are different facets of the same field Language Related Technologies R, NLP, DAR, ASR umans perform them well ifficult to specify algorithmically
Page 7
CSE 555: Sargur Srihari 7 Machine Learning rogramming computers to use example data or past experience Well-Posed Learning Problems computer program is said to learn from experience E ith respect to class of tasks and performance measure , f its performance at tasks , as measured by , improves with experience
Page 8
CSE 555: Sargur

Srihari 8 Classification Process (Decision as opposed to Inference)
Page 9
CSE 555: Sargur Srihari 9 Pattern Recognition Applications S ision ace/Handwriting/Hand peech peaker/Speech ouch aptics lfaction pple Ripe? TEXTUAL DATA Text Categorization nformation Retrieval ata Mining ntrusion Detection enome Sequence Matching
Page 10
CSE 555: Sargur Srihari 10 Pattern Recognition Processes bjects to be classified are sensed by transducer (camera) ignals are preprocessed eatures are extracted lassification is emitted
Page 11
CSE 555: Sargur Srihari 11 Generalization

Page 12
CSE 555: Sargur Srihari 12 Pattern Recognition System
Page 13
CSE 555: Sargur Srihari 13 Design Cycle
Page 14
CSE 555: Sargur Srihari 14 Document Recognition Applications ptical Character Recognition (OCR) andwriting Recognition riter Recognition
Page 15
CSE 555: Sargur Srihari 15 Writer Recognition Preprocessing Features Similarity
Page 16
CSE 555: Sargur Srihari 16 Address Interpretation Problem Pattern recognition tasks bject recognition (address vs non-address) wo-class discrimination (mp vs hw) ew class recognition (digits) olistic

vs analytical (words) ontextual-hmm(zip codes, words) any classes, but cataloged (postal directory)
Page 17
Contextual Information ountry/State/City IP Code treet Name rimary No (Street/PO Box ) econdary No (Apt) irm/Personal Name CSE 555: Sargur Srihari 17