In this paper we present a new architecture for combining classifiers. This approach integrates learning into the voting scheme used to aggregate individual classifiers decisions. This overcomes the drawbacks of havin...
详细信息
Accurate and automatic assessment of angiograms has been sought as a powerful diagnostic tool in medical analysis, such as in diabetic retinopathy. These retinal angiogram images are characterized by poor local contra...
详细信息
Accurate and automatic assessment of angiograms has been sought as a powerful diagnostic tool in medical analysis, such as in diabetic retinopathy. These retinal angiogram images are characterized by poor local contrast. Applications of existing edge detection algorithms yield unsatisfactory results. In this paper, a set of cascaded linear directional filters is used to better enhance edges. Results show improvement in visual quality of the images.
Cooperation by voting is one of the popular modular neural network decision-making strategies. Ensemble classifiers are multiple identical modules which use voting for post-learning classification. This paper suggests...
详细信息
Cooperation by voting is one of the popular modular neural network decision-making strategies. Ensemble classifiers are multiple identical modules which use voting for post-learning classification. This paper suggests a new cooperation scheme for ensembles which utilizes voting in the learning process itself. According to the suggested scheme, different modules would, automatically, focus on different regions in the input space. Hence, temporal crosstalk decreases and decision boundaries are drawn accurately in complex overlapping regions of the input space.
Hybrid continuous-time and discrete-event system models are developed for applications to intelligent robot and autonomous control systems. Certain general properties, including the interface structure between the dis...
详细信息
Hybrid continuous-time and discrete-event system models are developed for applications to intelligent robot and autonomous control systems. Certain general properties, including the interface structure between the discrete event part and the continuous part of hybrid systems, are studied. Petri nets are proposed as the model for the discrete event sub-system (DES), and state equations with a set of binary parameters as the model for the continuous time sub-system (CTS) in a hybrid system. The quantized CTS is formulated in prime event structures, which can also be modeled by a condition/event Petri net. Therefore, a complete hybrid system can be fit into an extended Petri net formalism, which provides a unified framework for hybrid system description. The validity of the proposed model is demonstrated by modeling a two-tank system.
暂无评论