Developing a real time algae detection platform using deep learning

Early detection of algal buildup by continuous monitoring of water bodies is a key component in combating this environmental hazard. Recent development in deep learning and object detection has open a new chapter in developing environmental monitoring methods. The field of machine learning have been used in numerous applications and in recent years and deep learning has become the leader in this domain. One of the primary differences between this model and typical machine learning model is the use of several layers to develop computational models. Traditional machine learning methods, on the other hand, necessitate manually designing features, which places a significant hardship on users. Deep learning can be defined as a machine learning representation learning algorithm based on large-scale data. This study, therefore, aims to explore implementation of object detection algorithms using deep learning for detection of algae. Developed models can be implemented on unmanned air vehicles (UAVs) and unmanned surface vehicles (USVs) and will perform the monitoring task. This model is cheap, easy to implement, and the results of the field experiment indicates that this method is reliable, and easy to implement.