Abstract: In the recording process of digital holographic measurement, the hologram is easily polluted by speckle noise, which may decrease the resolution of the hologram. In addition, the reconstructed effect is seriously affected by speckle noise in digital reconstruction. Thus it is important to study the filtering speckle algorithms for digital hologram. The median filtering algorithm, Lee filtering algorithm, Kuan filtering algorithm and SUSAN filtering algorithm were introduced to filter the speckle noise in hologram and reconstructed image. Then these algorithms were compared. The results showed that the SUSAN filtering algorithm was better in digital holographic technology. The speckle noises were suppressed significantly and the information of reconstructed images were well maintained.