As　a　newly　emerging　technology,deep　learning　is　a　very　promising　field　in　big　data　applications.Remote　sensing　applications　often　involve　huge　volume　data　obtained　daily　by　numerous　in-orbit　satellites.This　makes　it　a　perfect　area　for　data-driven　applications.Over　the　past　years,there　has　been　an　exponentially　increasing　interest　in　deep　learning　for　remote　sensing　image　processing,including　not　only　optical　imagery　but　also　synthetic　aperture　radar　(SAR)　imagery.In　addition　to　the　rapidly　growing　size　and　spectral,spatial　and　temporal　resolution　of　remote　sensing　data,there　are　other　challenges　that　are　unique　in　this　area,e.g.the　intrinsic　complexity　and　particularity　of　each　specific　sensor　and　their　multi-modality,the　fundamental　physical　properties　embedded　in　the　data,the　underlying　principles　for　information　retrieval,etc.To　promote　the　research　in　this　area,we　have　organized　a　special　focus　feature　on　deep　learning　in　remote　sensing　image　processing　in　the　SCIENCE　CHINA　Information　Sciences.