The large amounts of software repositories over the Internet are fundamentally changing the traditional paradigms of software maintenance. Efficient categorization of the massive projects for retrieving the relevant s...
详细信息
ISBN:
(纸本)9781467352185
The large amounts of software repositories over the Internet are fundamentally changing the traditional paradigms of software maintenance. Efficient categorization of the massive projects for retrieving the relevant software in these repositories is of vital importance for Internet-based maintenance tasks such as solution searching, best practices learning and so on. Many previous works have been conducted on software categorization by mining source code or byte code, which are only verified on relatively small collections of projects with coarse-grained categories or clusters. However, Internet-based software maintenance requires finer-grained, more scalable and language-independent categorization approaches. In this paper, we propose a novel approach to hierarchically categorize software projects based on their online profiles across multiple repositories. We design a SVM-based categorization framework to classify the massive number of software hierarchically. To improve the categorization performance, we aggregate different types of profile attributes from multiple repositories and design a weighted combination strategy which assigns greater weights to more important attributes. Extensive experiments are carried out on more than 18,000 projects across three repositories. The results show that our approach achieves significant improvements by using weighted combination, and the overall precision, recall and F-Measure can reach 71.41%, 65.60% and 68.38% in appropriate settings. Compared to the previous work, our approach presents competitive results with 123 finer-grained and multi-layered categories. In contrast to those using source code or byte code, our approach is more effective for large-scale and language-independent software categorization.
Cryo-electron microscopy (cryo-EM) has become a mainstream technology for solving spatial structures of biomacromolecules, while the processing of cryo-EM images is a very challenging task. One of the great challenges...
详细信息
This paper proposes a statistical method for no-reference image quality assessment using steerable pyramid decomposition without any prior knowledge about the distortions of the original image. Because the means of (l...
详细信息
To accommodate exponentially increasing traffic demands of vehicle-based applications, operators are utilizing offloading as a promising technique to improve quality of service (QoS), which gives rise to the applicati...
详细信息
The algorithm based on membrane system, also called membrane-inspired algorithm, has been shown to be powerful for solving combinatorial optimization problems, and it is increasingly used in practical engineering. Wit...
详细信息
Human interest input could greatly enhance the efficiency of the multi-robot exploration. However, most of the previous work did not take the intention of the operator into account. In this paper, a human interest ori...
详细信息
ISBN:
(纸本)9781479927456
Human interest input could greatly enhance the efficiency of the multi-robot exploration. However, most of the previous work did not take the intention of the operator into account. In this paper, a human interest oriented multi-robot exploration system is designed. The system is composed of a mobile base station and several robot explorers. A human interest oriented task allocation method is developed to enable these robot explorers to coordinate based on the operator's intention. Furthermore, an optimization index is proposed to balance the obedience, connectivity and explorability of the system; and the optimization problem is solved to generate the movements of the mobile base station and the robot explorers. Simulations and real-world experiments have demonstrate the effectiveness of the system and the efficiency of the exploration process.
Image defogging (IDF) removes influences of fogs from an image to improve its quality. Since defogged images can significantly boost the performance of subsequent processing, IDF has attracted many attentions from the...
详细信息
Image defogging (IDF) removes influences of fogs from an image to improve its quality. Since defogged images can significantly boost the performance of subsequent processing, IDF has attracted many attentions from the computer vision community. However, existing IDF algorithms are built on the assumption that light is scattered once by a grain. Since such assumption is violated if images are contaminated by dense haze or heavy fog, traditional IDF algorithms often fail in this situation. In this paper, we propose a hybrid image defogging (HIDF) algorithm to overcome this deficiency. In particular, HIDF applies the single scattering physics model (SSPM) to pixels dominated by single scattering of light, and applies the multiple scattering physics model (MSPM) to remaining pixels. To distinguish two types of pixels, HIDF utilizes the optical thickness of corresponding pixels. If optical thickness is smaller than a threshold that determines whether the single scattering or the multiple scattering dominates, HIDF applies the SSPM, and HIDF applies the MSPM otherwise. Experimental results on several popular foggy images demonstrate that HIDF competes with the state-of-the-art algorithms, and show the promise of HIDF for defogging heavily foggy images.
Opacity is an important information-flow security property that captures the plausible deniability for some "secret" of a system. In this paper, we investigate the problem of synthesizing controllers that en...
详细信息
ISBN:
(数字)9781728113982
ISBN:
(纸本)9781728113999
Opacity is an important information-flow security property that captures the plausible deniability for some "secret" of a system. In this paper, we investigate the problem of synthesizing controllers that enforce opacity for labeled transition systems (LTS). Most of the existing works on synthesis of opacity-enforcing controllers are based on the original system model, which may contain a large number of states. To mitigate the complexity of the controller synthesis procedure, we propose an abstraction-based approach for controller synthesis. Specifically, we propose notion of opacity-preserving alternating (bi)simulation relation for the purpose of abstraction. We show that, if the abstract system is opacity-preserving alternatingly simulated by the original system which may be significantly smaller, then we can synthesize an opacity-enforcing controller based on the abstract system and then refine it back to a controller enforcing opacity of the original system. We investigate both initial-state opacity and infinite-step opacity. We also show the effectiveness of the proposed approach by a set of examples.
In addressing the challenges of limited semantic labeling data and imprecise segmentation regions prevalent in the semantic segmentation of ship flame images, this study introduces a semi-supervised semantic segmentat...
In addressing the challenges of limited semantic labeling data and imprecise segmentation regions prevalent in the semantic segmentation of ship flame images, this study introduces a semi-supervised semantic segmentation method tailored for ship flame images. This approach is grounded in the principles of pairwise similarity with strong-weak perturbations. Adopting a pseudo-label self-learning strategy, the method employs weak-strong perturbation pairwise similarity, ensuring the propagation of high-confidence predictions during label generation, and thereby enhancing performance. Additionally, to tackle the inherent difficulties presented by smoke enveloping and obscuring flame regions in ship images, we have conceived a Vertical Continuity Feature Enhancement module. This module empowers the model to discern more explicit inter-class boundaries. Empirical evaluations on our proprietary dataset substantiate the superiority of our method, registering an improvement of 4.12 percentage points in mIoU over other semi-supervised approaches. Such results attest to the efficacy of the proposed pairwise similarity approach and the Vertical Continuity Feature Enhancement module in flame semantic segmentation.
In computer vision, the animation of objects has attracted a lot of attention, specially the animations of 3D face models. The animation of face models requires in general to manually adapt each generic movement (open...
详细信息
暂无评论