Confocal fluorescence microscopy provides a rapid method for acquiringhigh quality optically thin section images of wood suitable for meas-urement of cell dimensions. Single optical slice images of wood mayoccasionally contain artefacts due to differential light absorption causedby variation in the distance between the sample surface and the imagingplane across the field of view. Regional brightness variations, which wecall shading, may cause problems when such images are used for woodcell measurements using digital image analysis, affecting the accuracyof wood cell dimensions. We have compared various shading correctionmethods for confocal microscope images and investigated the effect ofshading on both the classification of cell wall pixels and the resulting celldimension measurements. Severe shading results in significant errors formeasurement of cell wall area, but smaller errors for cell wall thicknessand lumen diameter. Some shading correction methods have unwantedeffects on pixel classification and cell dimensions, while more effectivemethods remove the shading without introducing further artefacts. Theeffect of shading is influenced by choice of thresholding method.