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.
The software systems which are related to national science and technology projects are very crucial. This kind of systems always involves high technical factors and has to spend a large amount of money, so the quality...
详细信息
The software systems which are related to national science and technology projects are very crucial. This kind of systems always involves high technical factors and has to spend a large amount of money, so the quality and reliability of the software deserve to be further studied. Hence, we propose to apply four intelligent classification techniques most used in data mining fields, including Bayesian belief networks (BBN), nearest neighbor (NN), rough set (RS) and decision tree (DT), to validate the usefulness of software metrics for risk prediction. Results show that comparing with metrics such as Lines of code (LOC) and Cyclomatic complexity (V(G)) which are traditionally used for risk prediction, Halstead program difficulty (D), Number of executable statements (EXEC) and Halstead program volume (V) are the more effective metrics as risk predictors. By analyzing obtained results we also found that BBN was more effective than the other three methods in risk prediction.
Classification of multi-source remote sensing images has been studied for decades, and many methods have been proposed or improved. Most of these studies focus on how to improve the classifiers in order to obtain high...
详细信息
Classification of multi-source remote sensing images has been studied for decades, and many methods have been proposed or improved. Most of these studies focus on how to improve the classifiers in order to obtain higher classification accuracy. However, as we know, even if the most promising method such as neural network, its performance not only depends on the classifier itself, but also has relation with the training pattern (i.e. features). On consideration of this aspect, we propose an approach to feature selection and classification of multi-source remote sensing images based on Mallat fusion and residual error in this paper. Firstly, the fusion of multi-source images can provide a fused image which is more preferable for classification. And then a featureselection scheme approach based on fused image is proposed, which is to select effective subsets of features as inputs of a classifier by taking into account the residual error associated with each land-cover class. In addition, a classification technique base on selected features by using a feed-forward neural network is investigated. The results of computer experiments carried out on a multisource data set confirm the validity of the proposed approach.
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
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
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.
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
In this paper, a novel neural network based manifold learning method(NNBML)[1] recently appeared in the Journal of Science is introduced. It can effectively convert high-dimensional data into low-dimensional codes, wh...
详细信息
ISBN:
(纸本)9781601320438
In this paper, a novel neural network based manifold learning method(NNBML)[1] recently appeared in the Journal of Science is introduced. It can effectively convert high-dimensional data into low-dimensional codes, which are then used for classification. However, it performs not well while dealing with small size face database used for face recognition. We propose a solution generating more samples data based on the existing data. The proposed method is implemented on two well-known face databases, viz. ORL and Yale face databases. The experimental results show that NNBML is able to deal with the task of face recognition after more data samples generated using the proposed method, and also that NNBML outperforms LDA in terms of recognition rate.
暂无评论