Sand Castle Summation for Pixel Processor Arrays

Pixel Processor Arrays (PPA) present a new vision sensor/processor architecture consisting of a SIMD array of processor elements, each capable of light capture, storage, processing and local communication. Such a device allows visual data to be efficiently stored and manipulated directly upon the focal plane, but also demands the invention of new approaches and algorithms, suitable for the massively-parallel fine-grain processor arrays. In this paper we implement an image-wide population count algorithm exploiting the parallel processing of the PPA. Performing such a global count was previously unviable for vision processing tasks due to its exhaustive computation time. Our approach shows an improvement of typically two orders of magnitude reduction in computation time, thus allowing it to be incorporated as a core component of many vision tasks upon the PPA.