Detection of the number and volume of bubbles dissolved in a bioreactor with the help of machine vision

Document Type : Research Article


assistant professor of system and mechatronic group, new sciences and technologies, university of Tehran, Tehran, Iran


The purpose of this article is to design a bioreactor or fermenter equipped with image processing technology in real time with the ability to check the volume and The number of bubbles in the bioreactor container allows for instant and precise control for the growth of microorganisms.

In this bioreactor, optimal conditions are provided for the growth of microorganisms such as fungi, bacteria and yeast, and the culture of animal and plant cells is easily done in it. The use of this bioreactor allows microorganisms to grow for more than ten generations before being transferred to the production stage. In addition to controlling the volume of air bubbles, the parameters of temperature, CO2, engine speed, amount of oxygen and pH and other parameters are constantly being controlled.

In bioreactors, the measurement of bubbles in the container is very important. This importance is evident in the fact that it creates different generations of cells in the bioreactor. They are responsible for its production, which allows the proper growth and multiplication of microorganisms. In previous similar models, it is not possible to measure the amount of air in the container and control the homogeneity of the environment at the moment, which is eliminated with the help of this method.

In this article, in the first step, with the help of the camera, the images are processed moment by moment as the input data recorded in the next step, and in the last step, the processed images are created with the help of neural networks as output. The volume, quantity and dispersion ratio of the bubbles in the container and the simulation results show the efficiency of the proposed method.


Main Subjects