Automated classification of Bacterial Images extracted from Digital Microscope via Bag of Words Model

The performance recognition of bacteria cell images is an effective survey for treatment of various diseases caused by the bacteria. Many algorithms for bacteria classification are designed for the needs of analysis of large-scale microscopic image bacteria. However, the biologist interpretation is suffered from insufficient information and thus may lead to limited accuracy in the bacteria classification process. To handle this drawback, machine learning tools, and image analysis approaches tackled identification of different bacteria species for improving the clinical microbiology investigation. In the proposed study, 200 bacterial images for ten different bacteria species with 20 images for each specie are extracted from DIBaS (Digital Images of Bacteria Species dataset). This proposed framework is divided into image preprocessing phase which obtained by histogram equalization, feature extraction by Bag-of-words model and classification phase by Support Vector Machine (SVM). The main objective is to enhance the bacterial images and find the image feature descriptors from the enhanced images which allowing to classify the bacterial images. The experimental results provided an average accuracy of 97% with classifier speed for automated detection and classification of bacterial images which would greatly reduce the disease outbreaks in future researches.