一种新的基于地理信息的多媒体传感器网络路由协议

无线多媒体传感器网络作为一种全新的信息获取和处理技术,更多地关注于音频、视频、图像等大数据量、大信息量的多媒体数据的采集与处理,在军事、民用及商业领域中具有广阔的应用前景.传统的基于地理信息的路由协议并不适用于多媒体传感器网络.提出一种基于纯地理信息的路由协议,该协议调整网络节点的通信覆盖半径,然后利用改进的贪婪算法建立路由.仿真实验表明,该算法能够显著延长网络节点生存时间和延长网络整体生存时间.     

WSN中的一种多传感器数据融合算法

为了适应无线传感器网络资源受限的特点,提出一种多传感器数据融合算法,算法在传感层和网络层之间增加数据融合层,将采集的数据分为实时数据和非实时数据,对数据进行约筒处理,提高传感器节点传输被采集信息的效率.利用TinyOS进行仿真测试,结果表明:数据融合算法可以有效地减少网络中数据通讯流量,节省传感器节点能量,延长网络生存时间.     

面向声音监测的多媒体传感器节点硬件设计与实现

多媒体传感器网络能够采集和传输信息丰富的音频、视频、图像等多媒体信息,具有十分广泛的应用前景,是近年来无线传感器网络的研究热点。目前,国外多媒体传感器节点主要针对图像传输;国内使用的节点大多都难以满足多媒体信息处理和传输等方面的应用要求。本文针对鄱阳湖鸟类声音监测的应用,设计实现了一种新型的高性能多媒体传感器节点。实验结果表明,由该节点组成的多媒体传感器网络,能有效建立路由,实时地感知和采集网络覆盖区域内的多媒体信息。     

基于DCS的WMSN多视角视频编解码

为了有效解决无线多媒体传感器网络中多视角视频监控传输数据量大以及网络能量、资源受限的问题,提出了一种基于分布式压缩感知的高压缩率多视角视频编解码方法.对多视角视频序列进行分组处理,并将图像组分为关键帧和非关键帧;对关键帧采用基于压缩感知(compressed sensing,CS)的编解码方法进行处理;而在非关键帧的编码端采用联合稀疏表示方法对残差图像稀疏表示,解码端利用帧间时间相关性和多视角空间相关性预测生成当前视频帧,并借助差异补偿方法进一步提高预测准确性,同时提高了重构效果.实验结果表明,该方法取得较高的压缩率,重构出的图像质量比参考方法更高,且PSNR值得到了较大的提高.     

基于功能信任的无线传感器网络安全数据融合方法

针对无线传感器网络节点恶意行为和自身故障问题,将节点信任按其功能行为分为传感信任、融合信任和传递信任,提出一种基于功能信任的无线传感器网络安全数据融合方法,并利用局部相关一致性原理检测节点的传感功能行为.NS2环境下的仿真实验结果表明,该方法可以有效地避免恶意节点对数据融合结果造成的影响,提高了数据融合的安全性和可靠性,延长了网络的寿命.     

无线传感器网络信息时空融合模型与算法研究

信息融合是解决无线传感器网络能量和通信带宽受限的有效途径,提出了一种无线传感器网络信息时间和空间多级混合融合的结构模型。首先对距离汇聚节点较远的节点信息进行时间融合和对距离汇聚节点较近的节点信息在汇聚节点进行时空融合,然后在汇聚节点对两种节点的融合结果进行全网络的空间融合。就该模型提出了基于DS证据理论的多级时空融合算法。数据分析结果表明:该模型能够降低节点能耗,延长网络寿命。     

基于相关性感知的无线视频传感器网络路由协议

无线视频传感器网络（ WVSNs）节点的能量十分有限,为了降低传感器节点的能耗,延长网络生存时间,提出了一种基于相关性感知的视频传感器网络通信协议。首先对当前经典无线视频传感器网络路由协议的不足进行分析;然后从簇首节点的选择、簇的形成、节点感知方向的调整、簇内通信以及簇间通信等几个方面进行改进和优化;最后在Matlab 2012平台进行了仿真对比测试。仿真结果表明：该协议不仅能够延长无线视频传感器网络生命周期,而且提高了监测区域覆盖率。     

无线传感器网络节点实现模型

无线传感器网络是集成了嵌入系统、无线通信、分布计算、检测技术,由大量无线微型传感器节点组成的新型网络。无线传感器网络节点是构成无线传感器网络的基础。分析了无线传感网络节点的信号处理过程,对比研究了基于微处理器和基于现代信号处理在系统可编程的两种无线传感网络节点的实现模型。采用现代可编程实现的无线传感网络节点模型体积小、功耗低、速度快。基于微处理器的节点具有极大的灵活性。     

一种基于无线多媒体传感器网络的C4ISR视频侦察系统

C4ISR（指挥信息系统）在现代信息化作战中正显示出其强大的功能,而其中信息的获取又显得地位尤其突出。从指挥信息系统视频侦察系统的需求出发,利用WMSN（无线多媒体传感器网络）的自组织性、强壮性的特点,设计了一种基于无线多媒体传感器网络的C4ISR视频侦察系统,提出了其网络体系结构,给出了其节点和基站的设计方法和通信协议的实现。     

无线传感器网络节点实现模型

无线传感器网络是集成了嵌入系统、无线通信、分布计算、检测技术，由大量无线微型传感器节点组成的新型网络。无线传感器网络节点是构成无线传感器网络的基础。分析了无线传感网络节点的信号处理过程，对比研究了基于微处理器和基于现代信号处理在系统可编程的两种无线传感网络节点的实现模型。采用现代可编程实现的无线传感网络节点模型体积小、功耗低、速度快。基于微处理器的节点具有极大的灵活性。     

基于Android的多传感器信息融合技术应用

介绍基于Android的多传感器信息融合技术在气溶胶自动化检测中的应用。多个传感器采用统一接口方式组网,每一个独立传感器采集的数据通过蓝牙传输到网内的Android处理终端进行融合和分析。通过对5种传感器进行组网采集数据,结果证明多传感器信息融合技术的应用增强了信息处理系统的适应性,实现了传感器之间的数据互补,提高了系...     

Communication-resource-aware adaptive watermarking for multimedia authentication in wireless multimedia sensor networks

Efficient multimedia and specifically image authentication is critical and in demand to protect data vulnerability in wireless multimedia sensor networks (WMSN). This is to prevent malicious intruders from modifying and forging image contents over a network. Watermarking technique has been widely used to assert an image data authentication over wired networks; however, resource constraints (e.g. processing power, communication energy) in small sensors and the state of error-prone wireless channels result in fundamental challenges for developing efficient watermarking schemes in WMSN. These challenges include how to embed/protect/extract watermark efficiently and robustly in low-cost sensors and how to transmit authenticated image and multimedia with high energy efficiency. In this paper, we propose a communication-resource-aware and adaptive watermarking scheme for multimedia authentication in WMSN. Our contribution is two folds. First, the transmission quality for the watermark as well as watermarked multimedia authenticity by embedding watermark with adaptive coding redundancies, and by unequally allocating network resources to protect the image and multimedia packets with the watermark information. Second, communication energy efficiency and real time performance are achieved with the watermark being adaptive to the network condition and the processing delay reduced due to the exploited inter-frame correlation. The simulation and experimental results demonstrate that the proposed adaptive watermarking system can achieve considerable gains in terms of energy saving, image transmission quality and multimedia authentication performance.     

Adaptive haptic rendering for time‐varying haptic and video frame rates in multi‐modal interactions

In multi‐modal interactions including haptics, problems such as input sensor noise, temporal mismatch between graphics and haptics, and non‐constant refresh rates may cause non‐smooth force/torque display. This paper proposes temporal smoothing technique for haptic interaction using a sensing glove in multi‐modal applications. The proposed technique employs two processes: (1) a noise reduction method is applied to reduce jitter noise at the sensors in the sensing glove and (2) an adaptive force extrapolation is applied for time‐varying haptic and video frame rates. To demonstrate the performance of the proposed method, we developed a test platform to assess a simple box model and relatively complex models such as gamephone, portable media player (PMP). It was subsequently demonstrated that the proposed method can support smooth haptic interactions in multi‐modal applications where a haptic device and a sensing glove are used. Copyright © 2009 John Wiley & Sons, Ltd.     

Self-orienting wireless multimedia sensor networks for occlusion-free viewpoints

Wireless multimedia sensor networks (WMSN) are formations of a large number of compact form-factor computing devices that can capture multimedia content, such as video and audio, and communicate them over wireless channels. The efficiency of a WMSN heavily depends on the correct orientation (i.e., view) of its individual sensory units in the field. In this paper, we study the problem of self-orientation in WMSN, that is finding the most beneficial orientation for all multimedia sensors to maximize multimedia coverage. We propose a new algorithm to determine a node’s multimedia coverage and find the sensor orientation that minimizes the negative effect of occlusions and overlapping regions in the sensing field. Our approach enables multimedia sensor nodes to compute their directional coverage leading to an efficient and self-configurable sensor orientation calculation. By using simulations, we show that the occlusion-free viewpoint approach increases the multimedia coverage significantly. The self-orientation methodology is designed in the form of a distributed algorithm, making it a suitable candidate for deployment in practical systems.     

基于多尺度时域3D卷积的视频超分辨率重建

视频超分辨率是一项很有实用价值的工作.针对超高清产业中高分辨率资源较为匮乏的问题,为了有效利用视频序列帧间丰富的时间相关性信息及空间信息,提出一种基于多尺度时域3D卷积的视频超分辨率重建算法.该算法将输入的低分辨率视频序列帧分别通过不同时间尺度的3D卷积进行时空特征提取,3D卷积能够同时对空间与时间建模,相较于2D卷积...     

Ant-based routing for wireless multimedia sensor networks using multiple QoS metrics

In wireless sensor networks, most routing protocols consider energy savings as the main objective and assume data traffic with unconstrained delivery requirements to be a given. However, the introduction of video and imaging sensors unveils additional challenges. The transmission of video and imaging data requires both energy efficiency and QoS assurance (end-to-end delay and packet loss requirements), in order to ensure the efficient use of sensor resources as well as the integrity of the information collected. This paper presents a QoS routing model for Wireless Multimedia Sensor Networks (WMSN). Moreover, based on the traditional ant-based algorithm, an ant-based multi-QoS routing metric (AntSensNet) is proposed. The AntSensNet protocol builds a hierarchical structure on the network before choosing suitable paths to meet various QoS requirements from different kinds of traffic, thus maximizing network utilization, while improving its performance. In addition, AntSensNet is able to use a efficient multi-path video packet scheduling in order to get minimum video distortion transmission. Finally, extensive simulations are conducted to assess the effectiveness of this novel solution and a detailed discussion regarding the effects of different system parameters is provided. Compared to typical routing algorithms in sensor networks and the traditional ant-based algorithm, this new algorithm has better convergence and provides significantly better QoS for multiple types of services in wireless multimedia sensor networks.     

Quality assessment of perceptual color video based on a top-down framework and quaternion

Objective video quality assessment is of great importance in a variety of video processing applications. Most existing video quality metrics either focus primarily on capturing spatial artifacts in the video signal, or are designed to assess only grayscale video thereby ignoring important chrominance information. In this paper, on the basis of the top-down visual analysis of cognitive understanding and video features, we propose and develop a novel full-reference perceptual video assessment technique that accepts visual information inputs in the form of a quaternion consisting of contour, color and temporal information. Because of the more important role of chrominance information in the “border-to-surface” mechanism at early stages of cognitive visual processing, our new metric takes into account the chrominance information rather than the luminance information utilized in conventional video quality assessment. Our perceptual quaternion model employs singular value decomposition (SVD) and utilizes the human visual psychological features for SVD block weighting to better reflect perceptual focus and interest. Our major contributions include: a new perceptual quaternion that takes chrominance as one spatial feature, and temporal information to model motion or changes across adjacent frames; a three-level video quality measure to reflect visual psychology; and the two weighting methods based on entropy and frame correlation. Our experimental validation on the video quality experts’ group (VQEG) Phase I FR-TV test dataset demonstrated that our new assessment metric outperforms PSNR, SSIM, PVQM (P8) and has high correlation with perceived video quality.     

保留边缘与细节的压缩采样视频复原算法*

已有的压缩感知视频复原算法因过平滑效应难以保留视频帧的边缘与细节信息,对此提出一种基于混合稀疏性测量的压缩采样视频复原算法。编码端将视频序列分为关键帧与非关键帧,并使用相同的感知矩阵对帧的每块进行采样。解码端则设计了考虑局部稀疏性与全局稀疏性的混合稀疏性测量方案,并将其作为压缩感知视频复原问题的正则项;然后,通过分裂Bregman迭代算法对关键帧进行解码,并考虑视频帧间的时间相关性对非关键帧进行细化处理。基于多组仿真实验的结果表明,本算法获得了较好的视频复原精度,并具有理想的计算时间性能。     

结合面部动作单元感知的三维人脸重建算法

真实场景往往面临数据稀缺和数据动态变化的问题, 小样本增量学习的目的是利用少量数据推理数据知识并减缓模型对于旧知识的灾难性遗忘. 已有的小样本增量学习的算法(CEC和FACT等)主要是利用视觉特征来调整特征编码器或者分类器, 实现模型对于新数据的迁移和旧数据的抗遗忘. 但是少量数据的视觉特征往往难以建模一个类别的完整特征分布, 导致上述算法的泛化能力较弱. 相比于视觉特征, 图像类别描述的文本特征具有较好的泛化性和抗遗忘性. 因此, 在视觉语言模型的基础上, 研究基于文本知识嵌入的小样本增量学习, 通过在视觉特征中嵌入具有抗遗忘能力的文本特征, 实现小样本增量学习中新旧类别数据的有效学习. 具体而言, 在基础学习阶段, 利用视觉语言模型抽取图像的预训练视觉特征和类别的文本描述, 并通过文本编码器实现预训练视觉特征到文本空间的映射. 进一步利用视觉编码器融合学习到的文本特征和预训练视觉特征抽象具有高辨别能力的视觉特征. 在增量学习阶段, 提出类别空间引导的抗遗忘学习, 利用旧数据的类别空间编码和新数据特征微调视觉编码器和文本编码器, 实现新数据知识学习的同时复习旧知识. 在4个数据集(CIFAR-100, CUB-200, Car-196和 miniImageNet)上验证算法的有效性, 证明基于视觉语言模型文本知识嵌入可以在视觉特征的基础上进一步提升小样本增量学习的鲁棒性.     

基于K-SVD的低信噪比WMSN视频图像稀疏去噪

无线多媒体传感器网络WMSN因感知视频等信息的优势而被广泛应用,但受天气、光照等外因干扰,所采集视频图像常含有较为严重的噪声。因此,在低信噪比条件下进行视频图像去噪是保证WMSN视频监测有效性和可靠性的关键。在分析WMSN视频图像特征的基础上,首先对其进行周期性采集、分帧及帧差等预处理;然后对关键帧运用K SVD训练DCT冗余字典以充分稀疏表示图像特征,并采用基于残差比的改进型Batch OMP实现关键帧去噪及重构,而对残差帧则基于DCT冗余字典进行稀疏去噪处理;最后,叠加去噪后的关键帧和残差帧,从而整体上实现低信噪比WMSN视频图像的去噪及重构。实验表明,本算法能更加有效地、较为快速地滤除视频图像噪声,适用于低信噪比WMSN视频图像去噪。