Refinement is a necessary and effective step in some node localization schemes of wireless sensor networks (WSN). Suitable refinement procedure can improve the node localization accuracy and raise the robustness of th...
详细信息
ISBN:
(纸本)9780863418365
Refinement is a necessary and effective step in some node localization schemes of wireless sensor networks (WSN). Suitable refinement procedure can improve the node localization accuracy and raise the robustness of the localization algorithm. However, most existing refinement algorithms are costly duo to complex computation and frequent communication, and may induce serious coverage problem duo to nonconvergent iterations. In view of above facts, Steepest descent method is proposed to be used as the refinement algorithm in this paper, and corresponding simulation experiments are done to testify its feasibility and validity. The results show that steepest descent method can optimize the node positions to a fairish accuracy extent, and compared with existing refinement methods, it outperforms in communication cost, computation cost, and coverage rate.
Online learning is a very desirable capability for video-based algorithms. In this paper, we propose a novel framework to solve the problems of video-based face tracking and recognition by online updating twin GMMs. A...
详细信息
Pupil localization is a very important preprocessing step in many machine vision applications. Accurate and robust pupil localization especially in non-ideal eye images (such as images with defocusing, motion blur, oc...
详细信息
ISBN:
(纸本)9784901122078
Pupil localization is a very important preprocessing step in many machine vision applications. Accurate and robust pupil localization especially in non-ideal eye images (such as images with defocusing, motion blur, occlusion etc.) is a challenging task. In this paper, a detailed method to solve this problem is proposed. This method is implemented in three main steps: first, segment the rough pupil region based on Gaussian Mixture Model according to the gray level distribution of eye image;then modify the rough segmentation result using morphological method to minimize the influence of some disturbing factors;last step is to estimate the pupil parameters based on minimizing the least square error. The proposed method is first tested on CASIA iris image dataset, and then on our self-captured iris dataset which with more varieties. Experiments show that the proposed method can perform well for non-ideal eye images of various qualities.
Iris image quality assessment is an important part of iris recognition system because the qualities of iris images would largely influence the recognition results. In this paper, we analyze and compare several represe...
详细信息
ISBN:
(纸本)9784901122078
Iris image quality assessment is an important part of iris recognition system because the qualities of iris images would largely influence the recognition results. In this paper, we analyze and compare several representative quality assessment methods, and then propose an effective method based on Laplacian of Gaussian operator for iris image assessment. Through computer simulations of several typical algorithms on our iris image database, SJTU-IDB, the proposed method is shown superior to the compared quality assessment methods.
Two medieval Slavonic manuscripts are recorded, investigated and analyzed by philologists in collaboration with computer scientists. The aim of the project is to develop algorithms that support the philologists by aut...
详细信息
In face recognition, the dimensionality of raw data is very high, dimension reduction (Feature Extraction) should be applied before classification. There exist several feature extraction methods, commonly used are Pri...
详细信息
Schema matching is the task of finding semantic correspondences between elements of two schemas, which plays a key role in many database applications, such as data integration, electronic commerce, data warehouse, sem...
详细信息
Schema matching is the task of finding semantic correspondences between elements of two schemas, which plays a key role in many database applications, such as data integration, electronic commerce, data warehouse, semantic query processing, and XML message exchange, etc. Especially, it is a basic research issue in metadata management. Unfortunately, it still remains largely a manual, labor-intensive, and expensive process. In this paper, the schema matching problem is treated as a combinatorial problem. Firstly, schemas are transformed into multi-labeled graphs, which are the internal schema model for schema matching. Therefore, the schema matching problem is reduced to the labeled graph matching problem. Secondly, a generic graph similarity measure is discussed, which uses the labels of nodes and the edges to compute the similarity between the two schemas. Then, an objective function based on the multi-labeled graph similarity is proposed. Based on the objective function, a greedy matching algorithm is designed to find the desired matching state for schema matching. A prominent characteristic of this method is that the algorithm combines the feasible matching information to obtain optimal matching. Finally, some schema samples are used to test the greedy matching algorithm. The test results confirm that the algorithm is effective, which can obtain mapping results with high quality.
This paper presents a novel approach for point target detection of sea-clutter SAR images. Traditional methods for this application can be classified into two aspects: threshold segmentation based on intensity differe...
详细信息
This paper presents a novel approach for point target detection of sea-clutter SAR images. Traditional methods for this application can be classified into two aspects: threshold segmentation based on intensity difference and target extraction based on suitable denoising. However, they appear to be not effective enough especially when sea clutter is strong. Taking advantage of the essentials of both methods, a effective approach using space separation is developed based on fractal theory and independent component analysis. First, pointwise Holder exponent are computed and binary-fuzzy processing is used for enhancement;then, basis images and independent components of the processed image are respectively obtained by ICA technique. After that, according to separation criterion, the original space is separated into two subspaces called clean-space and noise-space with respective independent components and corresponding basis images. Finally, the recovery image is obtained after enhancing the independent components in clean-space. As the results show, the proposed method is validated and point target is extracted more efficiently compared with conventional ones.
ServiceBSP model is presented as an extension of BSP model with a view to the advantages of BSP model in Grid environment where large-scale and geographically distributed resources (abstracted as services) are availab...
详细信息
In this paper, we propose a scheme for moving object tracking from videos by combining mean shift and motion field statistics. For mean shift, we employ an enhanced spatial-range mean shift that enables a reduced numb...
详细信息
暂无评论