A low overhead video compression technique for energy-starved wireless multimedia sensor network
Tamal Pal, Sipra Das Bit · 2017
Development of wireless multimedia sensor network (WMSN) emphasizes the need for energy-efficient video transmission as one of the key features for a class of applications. In this work, WMSN nodes equipped with NCFA (New colour filter array), capture frames of video and then compress these in units of macroblocks. In inter-frame compression, exploiting human gaze direction macroblocks are selected adaptively by a newly proposed local histogram based parameter (LHBP) which is computed by the corresponding intensities and frequencies of two co-located macroblocks of input frames. Next to this, intra frame compression is applied on the selected macroblocks by employing an existing image compression by column dropping (ICCD) technique. Subsequently a newly proposed lightweight arithmetic coding is applied to reduce the content of the selected macroblocks to be transmitted. Performance of the scheme primarily is evaluated and compared through qualitative analyses. Finally the entire scheme is simulated in Cooja, the Contiki network simulator to make it readily implementable in real life mote e.g. MICAz. The results show that the proposed scheme saves on an average 40%-54% energy over the two state-of-the-art competing schemes while retaining acceptable reconstruction quality.