The problem related to wireless visual sensor network applied for possible applications such as habitat monitoring, border patrol, battlefield surveillance, agricultural monitoring, traffic monitoring, smart city network and so on where images have to be transmitted over a wireless medium is limited bandwidth, energy, on chip computational power and memory. The sensor nodes generally have a processor unit of low complexity with restricted power and random access memory within a variation of 10KB. The algorithms needed to transform and encode the images are computationally complicated and have high requirements as compared to the low power sensor nodes. We developed low power and low memory image coding algorithms and architectures for the applications of Wireless Sensor Networks providing scalablility and compression comparable to state-of-the-art algorithms. We also proposed a novel algorithm for forward multilevel transform which utilize only 24 Bytes of RAM for 5/3 Daubechies coefficients, independent of the image size. It enables implementation of image compression algorithms on Energy Efficient IC’s suitable for Remote monitoring and environment reconstruction in Visual Sensor Networks. The project was supported by Analog Devices India in their Anveshan Fellowship and was mentioned under Top 3 projects. The work is under review for publication.