Time filter

Source Type

Zhou W.,Zhejiang University of Science and Technology | Zhou W.,Zhejiang University | Yu L.,Zhejiang University | Yu L.,Zhejiang Provincial Key Laboratory of Information Network Technology
IEEE Transactions on Multimedia | Year: 2016

Perceptual quality assessment of distorted three-dimensional (3D) images has become a fundamental yet challenging issue in the field of 3D imaging. In this paper, we propose a general-purpose blind/no-reference (NR) 3D image quality assessment (IQA) metric that utilizes the complementary local patterns (the local magnitude pattern and the proposed generalized local directional pattern) of binocular energy response (BER) and binocular rivalry response (BRR). The main technical contribution of this research is that binocular visual perception and local structural distribution are considered for NR 3D-IQA. More specifically, the metric simulates the binocular visual perception using BER and BRR. Subsequently, the local patterns of the binocular responses' encoding maps are used to form various binocular quality-predictive features, which will change in the presence of distortions. After feature extraction, we use k-nearest neighbors-based machine learning to drive the overall quality score. We tested our proposed metric against two publicly available 3D databases; these tests confirm that the proposed metric's results consistently align with human subjective judgments. © 1999-2012 IEEE.

Hu Q.,Wenzhou University | Hu Q.,Zhejiang Provincial Key Laboratory of Information Network Technology | Tian Q.,De Zhou Vocational and Technical College | Tang Z.,Wenzhou University
International Journal of Wireless Information Networks | Year: 2013

Idle listening is one of the main factors for energy consumption in wireless sensor networks, and the duty cycle mechanism is widely used to reduce idle listening. In this paper, we present a new receiver-initiated duty cycling MAC protocol for wireless sensor networks, called reordering-passive MAC (RP-MAC), which includes receiver wake-up time estimation scheme and frame reordering scheme. We evaluate the performance of RP-MAC in ns-2 network simulator, and the simulation results shows that RP-MAC achieves higher energy efficiency, higher network throughput and lower end-to-end delay compared to other passive protocols, especially in case of heavy traffic or low duty cycle. © 2012 Springer Science+Business Media, LLC.

Li Y.,Zhejiang University of Technology | Li Y.,Zhejiang Provincial Key Laboratory of Information Network Technology | Shen Y.,Zhejiang University of Technology | Chi K.,Zhejiang University of Technology
KSII Transactions on Internet and Information Systems | Year: 2012

A tree routing structure is often adopted for many-to-one data gathering and aggregation in sensor networks. For real-time scenarios, considering lossy wireless links, it is an important issue how to construct a maximum-lifetime data gathering tree with delay constraint. In this work, we study the problem of lifetime-preserving and delay-constrained tree construction in unreliable sensor networks. We prove that the problem is NP-complete. A greedy approximation algorithm is proposed. We use expected transmissions count (ETX) as the link quality indicator, as well as a measure of delay. Our algorithm starts from an arbitrary least ETX tree, and iteratively adjusts the hierarchy of the tree to reduce the load on bottleneck nodes by pruning and grafting its sub-tree. The complexity of the proposed algorithm is O(N4). Finally, extensive simulations are carried out to verify our approach. Simulation results show that our algorithm provides longer lifetime in various situations compared to existing data gathering schemes. © 2012 KSII.

Xu N.,Zhejiang University | Huang A.,Zhejiang University | Huang A.,Zhejiang Provincial Key Laboratory of Information Network Technology | Hou T.-W.,National Cheng Kung University | Chen H.-H.,National Cheng Kung University
Wireless Communications and Mobile Computing | Year: 2012

One of the most challenging issues in wireless sensor networks is to meet the requirements of coverage and connectivity under given energy constraints. Most existing coverage and connectivity algorithms work to form tree networks when sensor nodes do not have location information of themselves. However, a tree topology network does not perform well in terms of energy efficiency and scalability if compared with a cluster network. In this paper, a novel topology control algorithm called Adaptive Random Clustering (ARC) is proposed to form a cluster network with required coverage and connectivity without location information. The performance of its coverage intensity and connectivity is analyzed based on the characteristics of cluster topology, and their proper parameters are determined. ARC inherits an excellent energy efficiency from cluster topology and avoids the collisions and overhearing of data packets. A good scalability can be achieved as only a limited number of channels are needed in ARC for a large-scale network. Furthermore, ARC can adjust the number of active nodes adaptively according to the required coverage to balance the energy consumption. Simulation results demonstrate that required coverage and connectivity can be satisfied and network lifetime is prolonged significantly. Copyright © 2010 John Wiley & Sons, Ltd.

Shen X.,Zhejiang University | Shen X.,Zhejiang Provincial Key Laboratory of Information Network Technology | Yu L.,Zhejiang University | Yu L.,Zhejiang Provincial Key Laboratory of Information Network Technology
Eurasip Journal on Image and Video Processing | Year: 2013

High efficiency video coding (HEVC) is the latest video coding standard that has been developed by JCT-VC. It employs plenty of efficient coding algorithms (e.g., highly flexible quad-tree coding block partitioning), and outperforms H.264/AVC by 35-43% bitrate reduction. However, it imposes enormous computational complexity on encoder due to the optimization processing in the efficient coding tools, especially the rate distortion optimization on coding unit (CU), prediction unit, and transform unit. In this article, we propose a CU splitting early termination algorithm to reduce the heavy computational burden on encoder. CU splitting is modeled as a binary classification problem, on which a support vector machine (SVM) is applied. In order to reduce the impact of outliers as well as to maintain the RD performance while a misclassification occurs, RD loss due to misclassification is introduced as weights in SVM training. Efficient and representative features are extracted and optimized by a wrapper approach to eliminate dependency on video content as well as on encoding configurations. Experimental results show that the proposed algorithm can achieve about 44.7% complexity reduction on average with only 1.35% BD-rate increase under the "random access" configuration, and 41.9% time saving with 1.66% BD-rate increase under the "low delay" setting, compared with the HEVC reference software. © 2013 Shen and Yu; licensee Springer.

Zhang J.,Zhejiang University | Zhang Z.,Zhejiang University | Zhang Z.,Zhejiang Provincial Key Laboratory of Information Network Technology | Luo H.,Zhejiang University | And 2 more authors.
International Journal of Communication Systems | Year: 2012

In this paper, an initial spectrum access control algorithm for cognitive wireless networks was studied. The objective was to maximize the number of admitted secondary users (SUs) under the constraints of interference temperature at multiple measurement points (MPs), while providing active SUs' QoS protection (AQP). Here, AQP means that the signal-to-interference-plus-noise ratios (SINRs) of all active SUs will not fluctuate below some predefined thresholds during a new SU's initial spectrum access process. To this end, we proposed a new distributive power control algorithm in conjunction with a simple alarm mechanism that ensures the interference temperatures at multiple MPs to be always below their thresholds during the initial spectrum access process. Then, two realizations of the spectrum access algorithms, namely, D-AQP and C-AQP, according to whether the MPs can act as local controllers or not, were presented. In the D-AQP algorithm, each SU updates its transmit power distributively based on the local information including the current SINR, the QoS requirement, and the indicator from the MPs so that the resultant SINRs of the active SUs are able to stay above the predefined thresholds. Whereas in the C-AQP algorithm, apart from the same procedure as mentioned, each MP is supposed to be able to decide which new SU should quit the initial spectrum access process so as to protect the communication quality of the primary system. Finally, the performances of the proposed algorithms were validated by extensive simulations. Copyright © 2011 John Wiley & Sons, Ltd.

Bao P.,Zhejiang University | Yu G.,Zhejiang University | Yu G.,Zhejiang Provincial Key Laboratory of Information Network Technology | Yin R.,Zhejiang University
2013 9th International Wireless Communications and Mobile Computing Conference, IWCMC 2013 | Year: 2013

In this paper, the interference mitigation problem in device-to-device (D2D) communication underlaying cellular networks using fractional frequency reuse (FFR) is studied. For such system, we propose a location based channel reusing scheme in which the D2D users in the inner region have a chance to reuse the channel resources of cellular users in the outer region of neighboring cells, and the D2D users in the outer region could reuse the channel resources of particular cellular users in the inner region of the same cell. Two novel concepts are introduced, namely the accessible region and the reusable region. To protect the outage probabilities of both cellular links and D2D links, only the D2D users in the accessible region could reuse the channel resources of cellular users in the reusable region. We present a specific method to calculate the boundary of the accessible region, as well as the reusable region. Simulation results demonstrate that the proposed scheme could effectively alleviate the interference between both users. © 2013 IEEE.

Xiang Z.,Zhejiang University | Xiang Z.,Zhejiang Provincial Key Laboratory of Information Network Technology | Sun B.,Zhejiang University | Sun B.,Zhejiang Provincial Key Laboratory of Information Network Technology
Guangxue Xuebao/Acta Optica Sinica | Year: 2012

Given the parameters of the mirror shape and the camera, a self-calibration method for catadioptric camera system is proposed to estimate the mirror pose with respect to the camera by using only one image containing no special designed calibration pattern. The ellipse corresponding to the mirror boundary in the image is used to compute possible mirror pose by planar circular posture estimation. With two pose candidates thereafter, the correct pose and the parameter of lens boundary simultaneously is found by matching the predicted and the observed image of lens boundary. The method features easy-operating, high accuracy and is suitable for non-single viewpoint system. Experimental results both on synthesized and real images demonstrate the validity of our method.

Zhang J.,Zhejiang University | Zhang J.,Zhejiang Provincial Key Laboratory of Information Network Technology | Zhang Z.,Zhejiang University | Zhang Z.,Zhejiang Provincial Key Laboratory of Information Network Technology
2011 IEEE Wireless Communications and Networking Conference, WCNC 2011 | Year: 2011

In this paper, the problem of initial link establishment in multi-channel Cognitive Radio Networks (CRN) is investigated. Different from the traditional scenario, in which the transceiver pair can establish an initial link through exchanging the control messages on the Common Control Channel (CCC), a more realistic scenario in the absence of CCC is considered in our work. Without the help of CCC, an initial link establishment approach based on the technique of group hopping is proposed and analyzed. In the asymptotic scenario N → ∈ (N: the number of available channels), we firstly derive the analytical expression of the condition, under which the initial link can be successfully established with probability arbitrarily close to 1. Secondly, we present the upper bound of average trial times and derive another condition, under which the initial link can be established within the required trial times in an average sense. Finally, we validate the analysis through extensive simulations. © 2011 IEEE.

Song J.,Ningbo University of Technology | Song J.,Zhejiang Provincial Key Laboratory of Information Network Technology | Wang L.,Ningbo University of Technology | Wang W.,Ningbo University of Technology
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012

Eyebrow is one of the most salient face features. It has a lot of potential applications in face recognition, nonverbal communication, and so on. In this paper, a novel eyebrow segmentation method based on binary edge image (BEI) is proposed. Our method firstly extracts BEI from a grayscale face image, and then connections between different face components in a BEI are removed using a specially designed algorithm. After that, some eyebrow-analogue segments are extracted from a BEI based on the geometrical property of eyebrows. The fourth step is to locate eyebrows using integral projection approach. Finally the perimeter of an eyebrow block is extracted to finish the segmentation of an eyebrow. Experimental results on a set of 517 AR images with different facial expression and illumination show that a correct eyebrow segmentation rate of 93.4% is achieved, indicating that the proposed method is robust to facial expression and illumination changes. © 2012 Springer-Verlag.

Loading Zhejiang Provincial Key Laboratory of Information Network Technology collaborators
Loading Zhejiang Provincial Key Laboratory of Information Network Technology collaborators