Distributed Image Compression for Image Sensor Networks

Abstract

As the demand for surveillance and monitoring continues to grow, the necessity for an efficient network to manage these tasks becomes increasingly critical. Wireless Multimedia Sensor Networks (WMSNs) effectively address these needs. However, the large volume of multimedia data poses several challenges for WMSNs, with energy consumption being the most significant issue. This work presents a solution to mitigate high energy consumption by implementing a hybrid image compression method distributed across sensors. This hybrid technique combines two lossy compression methods: discrete wavelet transform (DWT) and two-dimensional discrete cosine transform (DCT). Initially, the image is decomposed using DWT into low and high-frequency sub-bands. Each "n × n" block of the low-frequency subband

undergoes transformation using two-dimensional DCT. A fuzzy logic system (FLS) selects optimal sensors for the distribution tasks within the network. This approach significantly reduces processing time and extends the network's lifespan while maintaining data quality. Key work: WMSN, Energy Consumption, Image compression.