Being an essential concept in CAD/CAM and even in Concurrent Engineering, feature isthe theoretical basis of related technology, but formal definition of rt is still wanted. After reviewingprevious works , this paper ...Being an essential concept in CAD/CAM and even in Concurrent Engineering, feature isthe theoretical basis of related technology, but formal definition of rt is still wanted. After reviewingprevious works , this paper defines form feature as″a triplet(G ,D ,S) relevant to whole artifact or apart of it , where G,D,S are knowledge of geometry, manufacturing domain (CIM domain) and aspecific stage of product life cycle (CIM stage) , respectively. These knowledge together with aggre-gating and creating rules can be used for design , recognition and classification″ . In this definition .non-geometric semantics of form features are emphasized, and featured primitives, as feature con-structs, are defined by augmented adjacency graph (AAG) and geometric constraints. Finaliy. alge-braic specification is utilized to represent form features and illustrative examples are given.展开更多
A feature-centered scheme is presented to represent the feature-based variational partmodel. Constraint reasoning approach and numerical approach are trigged up to satisfy theconstraints between features. Coassociatio...A feature-centered scheme is presented to represent the feature-based variational partmodel. Constraint reasoning approach and numerical approach are trigged up to satisfy theconstraints between features. Coassociations between boundary elements in feature level and partlevel are maintained to implement a series of feature operations and to implement precision featureinodslling adaptive for paramatric part geometry model.GHCAD, a feature-based parametricmodelling system based on these ideas , is developed, and some examples are given finally.展开更多
This paper presents a variant of Haar-Iike feature used in Viola and Jones detection framework, called scattered rectangle feature, based on the common-component analysis of local region feature. Three common componen...This paper presents a variant of Haar-Iike feature used in Viola and Jones detection framework, called scattered rectangle feature, based on the common-component analysis of local region feature. Three common components, feature filter, feature structure and feature form, are extracted without concerning the details of the studied region features, which cast a new light on region feature design for specific applications and requirements： modifying some component（s） of a feature for an improved one or combining different components of existing features for a new favorable one. Scattered rectangle feature follows the former way, extending the feature structure component of Haar-like feature out of the restriction of the geometry adjacency rule, which results in a richer representation that explores much more orientations other than horizontal, vertical and diagonal, as well as misaligned, detached and non-rectangle shape information that is unreachable to Haar-Iike feature. The training result of the two face detectors in the experiments illustrates the benefits of scattered rectangle feature empirically; the comparison of the ROC curves under a rigid and objective detection criterion on MIT＋CMU upright face test set shows that the cascade based on scattered rectangle features outperforms that based on Haar-Iike features.展开更多
文摘Being an essential concept in CAD/CAM and even in Concurrent Engineering, feature isthe theoretical basis of related technology, but formal definition of rt is still wanted. After reviewingprevious works , this paper defines form feature as″a triplet(G ,D ,S) relevant to whole artifact or apart of it , where G,D,S are knowledge of geometry, manufacturing domain (CIM domain) and aspecific stage of product life cycle (CIM stage) , respectively. These knowledge together with aggre-gating and creating rules can be used for design , recognition and classification″ . In this definition .non-geometric semantics of form features are emphasized, and featured primitives, as feature con-structs, are defined by augmented adjacency graph (AAG) and geometric constraints. Finaliy. alge-braic specification is utilized to represent form features and illustrative examples are given.
文摘A feature-centered scheme is presented to represent the feature-based variational partmodel. Constraint reasoning approach and numerical approach are trigged up to satisfy theconstraints between features. Coassociations between boundary elements in feature level and partlevel are maintained to implement a series of feature operations and to implement precision featureinodslling adaptive for paramatric part geometry model.GHCAD, a feature-based parametricmodelling system based on these ideas , is developed, and some examples are given finally.
基金Supported by the National Basic Research Program of China (Grant No.2006CB303106)the Doctoral Subject Special Scientific Research Fund of the Ministry of Education of China (Grant No.20070335074)
文摘This paper presents a variant of Haar-Iike feature used in Viola and Jones detection framework, called scattered rectangle feature, based on the common-component analysis of local region feature. Three common components, feature filter, feature structure and feature form, are extracted without concerning the details of the studied region features, which cast a new light on region feature design for specific applications and requirements： modifying some component（s） of a feature for an improved one or combining different components of existing features for a new favorable one. Scattered rectangle feature follows the former way, extending the feature structure component of Haar-like feature out of the restriction of the geometry adjacency rule, which results in a richer representation that explores much more orientations other than horizontal, vertical and diagonal, as well as misaligned, detached and non-rectangle shape information that is unreachable to Haar-Iike feature. The training result of the two face detectors in the experiments illustrates the benefits of scattered rectangle feature empirically; the comparison of the ROC curves under a rigid and objective detection criterion on MIT＋CMU upright face test set shows that the cascade based on scattered rectangle features outperforms that based on Haar-Iike features.
