Query　suggestion　is　an　effective　human-computer　interaction　(HCI)　approach　in　information　retrieval.　According　to　human　vision　system,　the　image　retrieval　method　using　visual　query　suggestion　(VQS)　can　provide　a　friendly　query　interface　to　solve　the　query　ambiguity　problem.　In　this　paper,　the　content-based　image　retrieval　(CBIR)　system　is　realized　first.　Human-computer　interaction　used　VQS　to　obtain　users＇　intention.　In　this　step,　user　submitted　a　query　keyword　to　the　system.　VQS　is　utilized　to　provide　a　list　of　suggestions,　each　containing　a　keyword　and　a　collection　of　representative　images.　If　the　user　selects　one　of　the　image　suggestions,　this　image　will　be　viewed　as　the　key　image.　Then　CBIR　system　will　retrieve　the　image　sets　to　return　the　similar　images　based　on　the　similarity　of　image　features.　In　relevance　feedback,　user　scored　each　returned　image　by　the　slider　to　optimize　retrieval　results.　A　friendly　query　interface　is　designed　to　carry　out　HCI　in　our　system　and　the　experimental　result　shows　the　proposed　method　can　improve　the　average　recall　and　precision　efficiently.　©　2012　IEEE.