Digital　holography　is　particularly　well　suited　for　characterization　of　microstructure　such　as　surface　shape,　surface　nanostructures　and　surface　roughness.　The　direct　availability　of　both　amplitude　and　phase　information　offers　a　range　of　versatile　processing　techniques　that　can　be　applied　to　complex　field　data,　including　phase　imaging,　which　is　particularly　straightforward　in　digital　holography.　Based　on　digital　off-axis　lensless　Fourier　transform　holographic　configuration,　the　principle　of　topography　by　digital　holography　is　analyzed.　The　wavefront　deformation　created　by　numerical　Fresnel　reconstruction　is　then　studied.　Thus　the　model　of　phase　mask　which　is　a　parabolic　function　and　related　to　the　recording　parameters　is　built.　The　phase　mask　can　be　obtained　automatically　by　using　digital　hologram　itself　to　evaluate　automatically　and　accurately　the　values　of　the　parameters　involved　by　a　curve-fitting　procedures　applied　to　the　phase　data　extracted　along　two　profiles,　which　are　defined　in　the　region　where　sample　contributions　are　constant.　The　reconstructed　wave　field　can　be　simply　obtained　by　multiplying　the　phase　mask　with　Fourier　transform　of　the　hologram.　Both　theoretical　analysis　and　the　simulation　results　show　that　the　procedure　proposed　by　Colomb　is　more　suitable　for　phase　reconstruction　of　digital　off-axis　lensless　Fourier　transform　holography.　Moreover,　the　correcting　procedure　can　be　applied　to　compensate　high-order　aberrations.