To　improve　the　accuracy　of　leukocyte　segmentation,　this　paper　presents　a　novel　method　based　on　adaptive　histogram　thresholding　(AHT).　The　proposed　method　first　employs　color　component　combination　and　AHT　to　extract　the　nucleus　of　leukocyte　and　utilizes　image　color　features　to　remove　the　complex　backgrounds　such　as　red　blood　cells　(RBCs)　and　substantial　dyeing　impurities.　Then,　Canny　edge　detection　is　performed　to　extract　the　entire　leukocyte.　Finally,　the　cytoplasm　of　the　leukocyte　is　obtained　by　subtracting　the　nucleus　with　the　entire　leukocyte.　Experimental　results　on　an　image　dataset　containing　60　leukocyte　images　show　that　the　proposed　method　generates　more　accurate　segmentation　results　than　the　counterparts.　©　Springer　Nature　Singapore　Pte　Ltd.　2020.