The design and analysis of spatial data structures
The design and analysis of spatial data structures
A retrieval technique for similar shapes
SIGMOD '91 Proceedings of the 1991 ACM SIGMOD international conference on Management of data
Computing a shape's moments from its boundary
Pattern Recognition
Applications of parametric searching in geometric optimization
SODA '92 Proceedings of the third annual ACM-SIAM symposium on Discrete algorithms
Feature-Based Retrieval of Similar Shapes
Proceedings of the Ninth International Conference on Data Engineering
A Two-Stage Framework for Polygon Retrieval Using Minimum Circular Error Bound
ICIAP '97 Proceedings of the 9th International Conference on Image Analysis and Processing-Volume I - Volume I
Montage: An Image Database for the Fashion, Textile, and Clothing Industry in Hong Kong
ACCV '98 Proceedings of the Third Asian Conference on Computer Vision-Volume I - Volume I
BWT-based efficient shape matching
Proceedings of the 2007 ACM symposium on Applied computing
| Hi-index | 0.00 |
We propose a two-stage framework for polygonretrieval which incorporates both qualitative and quantitativemeasures of polygons in the first and second stage respectively. Thefirst stage uses Binary Shape Descriptor as a mean to prune thesearch space. The second stage uses any available polygon matchingand similarity measuring technique to compare model polygons with thetarget polygon. This two-stage framework uses a combination ofmodel-driven approach and data-driven approach. It is more efficientthan model-driven approach since it reduces the number of polygonsneeded to be compared. By using binary string as index, it alsoavoids the difficulty and inefficiency of manipulating complexmulti-dimensional index structure. This two-stage framework can beincorporated into image database systems for providing query-by-shapefacility. We also propose two similarity measures for polygons,namely Multi-Resolution Area Matching and Minimum Circular ErrorBound, which can be used in the second stage of the two-stageframework. We compare these two techniques with the HausdorffDistance method and the Normalized Coordinate System method. Ourexperiments show that Multi-Resolution Area Matching technique ismore efficient than the two methods and Minimum Circular Error Boundtechnique produces better polygon similarity measure than the twomethods.