Abstract:Due to the limitation of the nodes energy and computation ability in wireless visual sensor network (WVSN), the high complexity of traditional video coding in encoder is difficult to meet the requirement of practical application. Distributed source coding (DSC) can fully exploit the source statistics at the decoder to reduce the computational burden at the encoder. DSC provides a new way for video compression of orchard because it can effectively solve the problem of limited resources of sensor node in WVSN. Distributed video coding (DVC) is based on DSC, but it is a lossy compression coding technique. It suggests that under ideal conditions, the same rate distortion performance can be achieved as traditional video codecs. In practice however, there is still a significant performance gap between the two coding architectures. In order to improve the performance of DVC, a block adaptive DVC architecture was proposed according to the strong spatial and temporal correlation of the orchard video. It selected the coding pattern in accordance with the correlation of video content. WZ blocks were encoded with low density parity check (LDPC) code. For practical consideration of computational complexity and errorcorrection performance of the decoder, the jointbitplane LDPC decoding algorithm was used. Apple and grape orchard videos were tested and the experimental results show that the block adaptive DVC architecture can make the best use of the spatial and temporal correlation for compression coding. Compared with the traditional DVC, at the same quality of coding, it can save around 6%~10% and 9%~13% of the bit rate for apple video and grape video, respectively.