在资源受限的无线多媒体传感器网络中进行图像编码和传输,需要综合考虑能量消耗、压缩率和图像质量三者之间平衡的图像编码方案。研究无线多媒体传感器网络中视频监控图像序列的压缩,提出一种基于变化检测和改进JPEG算法的低复杂度图像编码方案。通过变化检测算法定位监控图像中的运动目标即感兴趣区域,仅传输兴趣区域以显著减少数据传输量,适应无线传感器节点存储转发能力有限的特点;改进JPEG算法中的DCT和量子化过程以降低计算复杂度,适应无线传感器节点计算能力有限的特点。算法复杂度分析和仿真实验结果表明,所提出的方法在保证应用所需图像质量的情况,有效地降低了无线传感器节点的数据通信量和计算过程能耗。
With resource-constrained wireless multimedia sensor networks,image coding and transmission must respect the trade-off among energy consumption,compression ratio and image quality.Compression of video-surveillance image sequences collected by a wireless multimedia sensor network is studied.A low-complexity image compression scheme based on change detection and adapted JPEG is proposed.For adaptation to the limited capacity of store and forward,change detection algorithm is used to locate the region of interest and cut down the data for transmission by only transmitting the region of interest.For adaptation to the limited computational capability,DCT and quantization of JPEG is optimized to reduce the computation complexity.Computation complexity analysis and simulation results indicate that the proposed image compression scheme effectively reduces the data traffic and energy consumption of computation under the required image quality.