Localization　of　the　optic　disc　(OD)　and　the　fovea　is　a　crucial　task　towards　developing　automatic　diagnosis　and　screening　tools　for　retinal　disease.　The　aim　of　the　paper　was　to　propose　a　simplified　method　for　detecting　the　both　features　jointly.　On　the　basis　of　directional　local　contrast　(DLC)　technique,　the　fovea　and　OD　candidates　were　detected　successively,　then　position　constraint　information　between　them　was　used　to　determine　the　true　features.　The　proposed　method　was　evaluated　using　publicly　available　databases　MESSIDOR　(containing　1200　images)　and　HEI-MED　(containing　169　images).　The　experiment　showed　that　a　good　success　rates　of　97.41%　and　97.33%　for　the　fovea　and　OD　detection　can　be　achieved　on　the　MESSIDOR　database,　respectively,　and　98.8%　for　both　on　the　HEI-MED　dataset.　In　terms　of　the　fovea　localization　accuracy,　by　analyzing　the　1136　cases　among　MESSIDOR　with　ground-truth　fovea　position,　the　method　could　provide　a　fovea　center　pixel　categorized　within　R/8,　R/4,　R/2　and　1R　criteria　in　60.39%,　91.36%,　98.32%,　and　99.03%　of　the　cases,　respectively.　Overall,　the　proposed　method　can　be　a　feasible　tool　for　eye　diseases　screening　or　assisted　diagnosis.