Cluster analysis is an un-supervised learning technique that is widely used in the process of topic discovery from text. The research presented here proposes a novel un-supervised learning approach based on aggregatio...
详细信息
This paper presents an algorithm for extraction of phrases from text *** e algorithm builds phrases by iteratively merging bigrams according to an association *** o association measures are presented: mutual informati...
详细信息
In this paper, we combine a novel set of image features called virtual circles with edge direction, to provide an efficient alignment algorithm for image registration under similarity transformations. Virtual circles ...
详细信息
In this paper we combine a novel set of image features called virtual circles with edge direction, to provide an efficient alignment algorithm for image registration under similarity transformations. Virtual circles c...
详细信息
In this paper we combine a novel set of image features called virtual circles with edge direction, to provide an efficient alignment algorithm for image registration under similarity transformations. Virtual circles can be extracted efficiently from many types of images with the help of the distance transform. They can also have extra information, such as their radius and the dominant edge direction in the neighborhood, which can be used for efficient registration. An heuristic called the smoothness criterion is used to choose virtual circles which are less likely to be corrupted by background noise, and to increases the efficiency of the algorithm. Experimental results have shown that this method is highly automatic and has linear complexity in terms of the number of virtual circles extracted.
This paper presents a multi-agent system for coordinating the deployment of multiple sensors in a modeled environment. The sensing task is the maximal sensor coverage of one or more targets in a scene and the position...
详细信息
This paper presents a multi-agent system for coordinating the deployment of multiple sensors in a modeled environment. The sensing task is the maximal sensor coverage of one or more targets in a scene and the position of each sensor is controlled by an autonomous agent. The agents rely on negotiation to achieve the level of coordination necessary to accomplish the given sensing task. Currently, the system focuses on the use of cameras for visual inspection tasks in which a single camera may be inadequate due to occluding objects in the scene, the number of targets to be observed or the sheer size of the target. The paper presents the negotiation mechanism developed for the multi-agent planning of the camera positions and illustrates its effectiveness by an example. Results show that the agents were able to autonomously position the cameras so as to allow for an acceptable level of coverage of the targets being observed.
There exist numerous schemes and methods to determine the output of an ensemble of classifiers. The most common approach being the majority vote. Furthermore, we might expect that an improvement can be achieved if the...
详细信息
There exist numerous schemes and methods to determine the output of an ensemble of classifiers. The most common approach being the majority vote. Furthermore, we might expect that an improvement can be achieved if there is a method by which we may weigh the members of the ensemble according to their individual performance. The feature based approach presented an architecture that tries to approach this target. However, if there is a way that the final classification may influence these weights we should expect an increased performance in the overall classification task. In this paper we present a new training algorithm that utilizes a feedback mechanism to iteratively improve the classification capability of the feature based approach. This approach is compared with the standard training method as well as standard aggregation schemes for combining classifier ensembles. Empirical results show that this architecture improved on classification accuracy.
In this paper, we combine a novel set of image features called virtual circles with edge direction, to provide an efficient alignment algorithm for image registration under similarity transformations. Virtual circles ...
详细信息
In this paper, a new graph data structure for 2-D shape representation is proposed. The new structure is called a concavity graph, and is an evolution from the already known "concavity tree". Even though a c...
详细信息
In this paper, a new graph data structure for 2-D shape representation is proposed. The new structure is called a concavity graph, and is an evolution from the already known "concavity tree". Even though a c...
详细信息
In this paper, a new graph data structure for 2-D shape representation is proposed. The new structure is called a concavity graph, and is an evolution from the already known "concavity tree". Even though a concavity graph bears a fundamental resemblance to a concavity tree, the former is able to describe the shape of multiple objects in an image and their spatial configuration, and is hence inherently more complex. The aim of concavity graphs is two-fold: first we want to analyze the patterns in a multi-object image in a way that will (1) provide better representation of their shapes, and (2) convey useful information about how they "interact" together. Second, we want our analysis technique to facilitate similarity matching between two images. This paper introduces the new structure and outlines how it can be used for shape representation as well as similarity matching.
In this paper, a software architecture for multiple autonomous agents in a real-time environment is described. The software architecture for autonomous agents must effectively connect perception modules, planning modu...
详细信息
In this paper, a software architecture for multiple autonomous agents in a real-time environment is described. The software architecture for autonomous agents must effectively connect perception modules, planning modules and action modules. It is necessary to define both an interface and a data flow among different modules. The proposed software architecture connects reactive modules with deliberative modules. The architecture proved its feasibility by controlling a prototype of multiple model digital trains.
暂无评论