Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

12

TIP
1998

favoriteEmaildiscussreport

111views more TIP 1998»

Texture information in run-length matrices

13 years 4 months ago

Texture information in run-length matrices

Download mmlab.ie.cuhk.edu.hk

Abstract—We use a multilevel dominant eigenvector estimation algorithm to develop a new run-length texture feature extraction algorithm that preserves much of the texture information in run-length matrices and signiﬁcantly improves image classiﬁcation accuracy over traditional run-length techniques. The advantage of this approach is demonstrated experimentally by the classiﬁcation of two texture data sets. Comparisons with other methods demonstrate that the run-length matrices contain great discriminatory information and that a good method of extracting such information is of paramount importance to successful classiﬁcation.

Xiaoou Tang

Real-time Traffic

Run-length | Run-length Matrices | TIP 1998 | Traditional Run-length Techniques |

claim paper

Related Content

» Spatial Texture Analysis A Comparative Study

» Using covariance matrices for unsupervised texture segmentation

» Texture Image Retrieval Using Novel Nonseparable Filter Banks Based on Centrally Symmetric...

» Extracting Boundaries from Images by Comparing Cooccurrence Matrices

» Adaptive Concept Learning through Clustering and Aggregation of Relational Data

» Color image segmentation based on three levels of texture statistical evaluation

» Retrieving scale from quasistationary images

» MultiView Image Registration for WideBaseline Visual Sensor Networks

» Improving LSB steganalysis using marginal and joint probabilistic distributions

Post Info
More Details (n/a)

Added	23 Dec 2010
Updated	23 Dec 2010
Type	Journal
Year	1998
Where	TIP
Authors	Xiaoou Tang

Comments (0)