Sketch-Based shape retrieval using length and curvature of 2d digital contours

Authors:
Abdolah Chalechale;Golshah Naghdy;Prashan Premaratne
Affiliations:
School of Electrical, Computer & Telecommunications Engineering, University of Wollongong, NSW, Australia;School of Electrical, Computer & Telecommunications Engineering, University of Wollongong, NSW, Australia;School of Electrical, Computer & Telecommunications Engineering, University of Wollongong, NSW, Australia
Venue:
IWCIA'04 Proceedings of the 10th international conference on Combinatorial Image Analysis
Year:
2004

Citing 9
Cited 0

An Experimental Comparison of Autoregressive and Fourier-Based Descriptors in 2D Shape Classification

IEEE Transactions on Pattern Analysis and Machine Intelligence
VisualSEEk: a fully automated content-based image query system

MULTIMEDIA '96 Proceedings of the fourth ACM international conference on Multimedia
Digital Image Processing

Digital Image Processing
Affine-Invariant Sketch-Based Retrieval of Images

CGI '01 Computer Graphics International 2001
Sketch-Based Images Database Retrieval

MIS '98 Proceedings of the 4th International Workshop on Advances in Multimedia Information Systems
Evaluation of MPEG-7 shape descriptors against other shape descriptors

Multimedia Systems
Sketch-based image matching Using Angular partitioning

IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans
A contour-based approach to multisensor image registration

IEEE Transactions on Image Processing
MPEG-7 visual shape descriptors

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a novel effective method for line segment extraction using chain code differentiation. The resulting line segments are employed for shape feature extraction. Length distribution of the extracted segments along with distribution of the angle between adjacent segments are exploited to extract compact hybrid features. The extracted features are used for sketch-based shape retrieval. Comparative results obtained from six other well known methods within the literature have been discussed. Using MPEG-7 contour shape database (CE-1) as the test bed, the new proposed method shows significant improvement in retrieval performance for sketch-based shape retrieval. The Average Normalized Modified Retrieval Rank (ANMRR) is used as the performance indicator. Although the retrieval performance has been improved using the proposed method, its computational intensity and subsequently, its feature extraction time are slightly higher than some other methods.