Cervical virtual slides are ultra-large, can have size up to 120K x 80K pixels. This paper introduces an image segmentation method for the automated identification of Squamous epithelium from such virtual slides. In order to produce the best segmentation results, in addition to saving processing time and memory, a multiresolution segmentation strategy was developed. The Squamous epithelium layer is first segmented at a low resolution (2X magnification). The boundaries of segmented Squamous epithelium are further fine tuned at the highest resolution of 40X magnification, using an iterative boundary expanding-shrinking method. The block-based segmentation method uses robust texture feature vectors in combination with a Support Vector Machine (SVM) to perform classification. Medical histology rules are finally applied to remove misclassifications. Results demonstrate that, with typical virtual slides, classification accuracies of between 94.9% and 96.3% are achieved.