Towards a provenance framework for sub-image processing for astronomical data

While there has been advances in observational equipment that generate huge high quality images, the processing of these images remains a major bottleneck. We show that provenance data collected during the processing of data can be reused to perform selective processing of data and support network collaboration without clogging distribution networks. We introduce the idea of sub-image processing (SIMP) in the context of processing a subset of pixels of an image and the use of provenance data to assemble pipelines and to select processing metadata for SIMP. We describe an implementation of SIMP in Astro-WISE