Thispaperproposesalearning-basedmethodfortextdetectionandtextsegmentationinnaturalsceneimages.First,theinputimageisdecomposedintomultipleconnected-components(CCs)byNiblackclusteringalgorithm.ThenalltheCCsincludingtextCCsandnon-textCCsareverifiedontheirtextfeaturesbya2-stageclassificationmodule,wheremostnon-textCCsarediscardedbyanattentionalcascadeclassifierandremainingCCsarefurtherverifiedbyanSVM.AlltheacceptedCCsareoutputtoresultintextonlybinaryimage.Experimentswithmanyimagesindifferentscenesshowedsatisfactoryperformanceofourproposedmethod.