Count of bacteria and yeast in microbial bioproduct using digital image processing

The count of microorganisms in substances from different industries, like the count of bacteria and yeasts, is a necessary and important process since long time ago. Traditionally, in the industries this process is performed by experts observing the samples in the microscopes, which is time-consuming and varies depending on the degree of expertise of the experts. Currently, the use of digital images of the samples to be analyzed is a variant widely used for such count task. In that sense, several methods have been created in recent years to make this process, but none of them covers the wide range of diversity that can be found in the real microbiological world. With these ideas as premises, a new method for count bacteria and yeasts in microbial bioproducts using digital images is presented in this paper, in order to provide to experts the approximate number of those microorganism. The method involves basic operations of digital image processing like contour detection, morphological operations and statistical analysis; and it was developed in Python language using the OpenCV library. The results obtained were evaluated by microbiological experts proved to have an acceptable performance for the context of use.