Scribd
Upload
41 views17 pages

Feature Extraction: Dr. Mallikarjun Hangarge

This document discusses different types of image features that can be extracted for purposes like image recognition. It outlines two main types: global features and local features. Global features look at overall image properties but lose spatial information, while local features capture properties in local regions to support partial matching while being robust to brightness changes. The document then discusses various applications of feature extraction and recognition like traffic sign recognition, texture recognition using texton dictionaries, and bag-of-features models for object recognition.

Uploaded by

Dr.Mallikarjun Hangarge
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
41 views17 pages

Feature Extraction: Dr. Mallikarjun Hangarge

This document discusses different types of image features that can be extracted for purposes like image recognition. It outlines two main types: global features and local features. Global features look at overall image properties but lose spatial information, while local features capture properties in local regions to support partial matching while being robust to brightness changes. The document then discusses various applications of feature extraction and recognition like traffic sign recognition, texture recognition using texton dictionaries, and bag-of-features models for object recognition.

Uploaded by

Dr.Mallikarjun Hangarge
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 17

Feature Extraction

Dr. Mallikarjun Hangarge


Types of Image Features
• Global features – global properties of an image,
including intensity histogram, frequency domain
descriptors, covariance matrix and high order
statistics, etc.

Sensitive to brightness
Loss of spatial information
No support to partial matching
High dimensionality
4/8/2018 2
Local features – local regions with special properties,
including edges, corners, lines, curves, regions with
special properties, etc.

Ignores the small brightness change


Captures spatial information
Support to partial matching
High dimensionality
Recognition
• The action or the process of recognizing or
being recognized.
Recognition

Re + Cognition

Learning Labeling
4/8/2018 DSIP_SRTM University, Latur 4
Text detection & recognition or removal

Trafic sign recognition Calcite localization/assessment Detect tissue on the slide


License plate recognition
Single object images; defect detection on the object
Bag-of-features models

Many slides adapted from Fei-Fei Li, Rob Fergus, and Antonio Torralba
Texture recognition
• Texture is characterized by the repetition of basic
elements or textons
Texture recognition
histogram

Universal texton dictionary

Julesz, 1981; Cula & Dana, 2001; Leung & Malik 2001; Mori, Belongie & Malik, 2001; Schmid
2001; Varma & Zisserman, 2002, 2003; Lazebnik, Schmid & Ponce, 2003
Bags of features for object recognition

face, flowers, building

• Works pretty well for image-level classification

Csurka et al. (2004), Willamowski et al. (2005), Grauman & Darrell (2005), Sivic et al. (2003, 2005)
Classification
• Basic idea:
– If it walks like a duck, quacks like a duck, then it’s probably
a duck
Compute
Distance Test Record

Training Choose k of the


Records “nearest” records

12
K- Nearest Neighbor

X X X

(a) 1-nearest neighbor (b) 2-nearest neighbor (c) 3-nearest neighbor


K-nearest neighbors of a record x are data points that
have the k smallest distance to x

13
Case Studies

Word Retrieval from Kannada Documents


 Elsevier (Demo)
 IEEE
Handwritten Script Identification

•ICFHR- 2014, Greece

•ICDAR- 2013, USA


Acknowledgement

We are thankful to Prof. Vikas Humbe for giving us


an opportunity to share our research experience
International Conference on Image
Processing and Pattern Recognition
Dec 16-17, 2016

www.rtip2r.org
....................... ?

You might also like