Dyed barley cells in microscope colour images of biological experiments are analysed for the occurrence of haustoria of the powdery mildew fungus by a fully automated screening system. The region of interest in the images is found by applying Canny's edge detector to the hue channel of the HSV colour space. For the segmentation of potential haustoria within the dyed cells, two different methods are considered: A clustering in RGB colour space using the Expectation Maximisation (EM) algorithm, and morphological contrast enhancement of the colour image with subsequent hysteresis thresholding in the saturation channel of the enhanced images. The second approach seems to be more viable because of its robustness and more promising results.