Aspergillus can cause disease known as aspergillosis which usually infected human beings and needs a proper diagnosis method for further treatment. The most popular method that has been used by researcher is direct microscopic. However, this method still has high potential to cause some flaws related to diagnosis accuracy and fungal sensitization. Therefore, an image processing is approach to overcome the problem. In this paper, a combination of image enhancement and segmentation techniques is proposed. An image thresholding, mathematical morphological, active region based segmentation and image fusion is employed to improve the quality of particular area which gives the most important information of fungi. The image is segmented with three different conditions which are single masking for single subject, single masking for multiple subjects and double masking for multiple subjects in order to determine the effectiveness of proposed method. Experimental evaluation based on quality of image shows that the single masking for multiple subjects has better performance with 25.3235531 dB of PSNR. This shows that the proposed technique is able to enhance and segmented the images effectively.
