Because　of　the　large　difference　of　the　densities　in　the　lung　and　other　structures,　the　chest　x-ray　image　behaves　as　a　wide-range　intensity　distribution,　which　brings　on　a　bit　of　difficulty　to　investigate　the　focus.　In　the　paper,　according　to　the　intensity　properties　of　the　chest　radiograph,　the　chest　radiographic　image　is　divided　into　three　subregions,　and　a　piecewise　linear　transformation　model　is　established.　An　approach　of　automatic　enhancement　is　presented,　based　on　the　gray-level　normalization.　The　average　enhanced　ratios　of　three　subregions　of　the　normal　and　severe　acute　respiratory　syndrome　image　are　increased　by　10.70%　and　25.55%,　respectively.　The　technique　is　proved　to　be　effective　through　the　evaluation　of　the　improved　images.