80.7%。Characteristics of the Aphis gossypii area, background and mixed area (include Aphis gossypii area and background) were analyzed and principle of determining was establish based on the threshold G component. Then Aphis gossypii area and non-Aphis gossypii area were separated using the threshold G component. For the overlapping Aphis gossypii, the input image were marked using the minimum extension transform, then distance transform and watershed algorithm was applied to the marked image, and the overlapping was removed. Experimental results showed that this algorithm could effectively segment the overlapping Aphis gossypii. The sum of over-segmentation rate and under-segmentation rate was 3.14%. The accurate rate was 96.2%, which was higher than the direct counting.