How　to　use　low-level　image　features　such　as　color,　texture,　shape,　spatial　relationship　etc.　logically　in　retrieval　is　an　important　problem　for　both　retrieval　system　and　users.　An　image　retrieval　method　based　on　the　four　kinds　of　image　low-level　features　is　proposed　in　this　paper.　According　to　image　texture　characteristic,　a　kind　of　image　feature　statistic　is　defined.　By　using　feature　weight　assignment　operators　designed　here,　the　method　can　assign　weight　to　the　above　four　features　adaptively　and　realize　image　retrieval　based　on　combined　image　features.　The　experiment　results　show　that　the　method　mentioned　above　is　more　efficient　than　those　based　on　only　one　kind　visual　feature.　©　2007　IEEE.