Image　compression　plays　more　and　more　important　role　in　image　processing.　Image　sparse　coding　with　learned　over-complete　dictionaries　shows　promising　results　on　image　compression　by　representing　images　with　dictionary　atoms　compactly.　Within　the　sparse　coding　based　compression　framework,　a　sparse　dictionary　is　first　learned　from　training　images　in　a　predefined　image　library,　and　then　an　image　is　compressed　by　representing　its　non-overlapping　image　patches　as　linear　combination　of　very　few　dictionary　atoms,　which　is　called　sparse　coding.　In　this　paper,　we　proposed　a　content　adaptive　sparse　dictionary　for　image　set　compression　based　on　sparse　coding.　For　a　set　of　similar　images　to　be　compressed,　first　we　divided　image　patches　into　DC　and　AC　components.　For　the　AC　components,　a　clustering　algorithm　is　used　to　get　cluster　centers.　Then　a　content　adaptive　dictionary　will　be　learned　according　to　each　cluster　center.　We　compared　our　method　with　RLS-DLA　method　and　JPEG　method　to　validate　the　performance　of　our　method,　and　experimental　results　show　that　our　method　outperforms　the　comparing　methods　at　high　bitrate.