A New Video Compression Scheme Combining Conditional Replenishment and Address Event Representation

A novel video compression scheme is presented in this paper. This algorithm is implemented based on a conditional replenishment Address Event Representation (AER) imager, where image signals are compressed on the sensor level by employing the conditional replenishment algorithm. The intensity value of each pixel is compared to the last replenished frame. If the magnitude of the difference exceeds a certain threshold, the pixel value is quantized by 1-bit Fast Boundary Adaptation Rule (FBAR), while the address of the pixel is located and read-out by the AER method. Simulation result shows that a compression ratio of 11.73 can be achieved while obtaining a PSNR value of 35.22dB. Bandwidth requirement is reduced since pixel intensity levels are ordered before applying 1-bit adaptive quantization. Simulation was also performed on the raster scanning read-out case for the sake of comparison.