Automatic Feature Extraction Method Research Articles

Automated Feature Generation in Large-Scale Geospatial Libraries for Content-Based Indexing

We describe a method for indexing and retrieving high-resolution image regions in large geospatial data libraries. An automated feature extraction method is used that generates a unique and specific structural description of each segment of a tessellated input image file. These tessellated regions are then merged into similar groups, or sub-regions, and indexed to provide flexible and varied retrieval in a query-by-example environment. The methods of tessellation, feature extraction, sub-region clustering, indexing, and retrieval are described and demonstrated using a geospatial library representing a 153 km2 region of land in East Tennessee at 0.5 m per pixel resolution.

Automatic extraction of form features for casting

This paper details a new method for automatic feature extraction from a Boundary Representation of the solid model. The proposed method is based principally on the evaluation of two topological invariants of concave parts of the modeled object. The result is a decomposition of the object boundary into surface components which represent shape features. Some aspects of the topology of bordered surfaces are discussed. The feature extraction method has been optimized for automatic core pattern design and advice in casting design. The main aspects of the “domain of application” of the method are considered, and six categories of shape features identified. The extraction rules proposed for a specific domain of application could be extended in different feature recognition contexts.