A Comparative Analysis of Machine Learning Methods for Detecting Microalgal Outbreak using Remote Sensing Images

Abstract

The algae formation in the water bodies has been increased rapidly to demolish the algae present in the water the proper information should be retrieved from the water bodies that information can be attained from the machine learning methods and remote sensing algorithms that is used in below scenario . Due to complex composition of  farther detecting scenes inaccessible detecting picture scene classification is still a challenging errand . By the remote sensing algorithm the images that are inside the water bodies are captured and they are further processed for the preventive measurements. Also the types of algae are also classified in many forms that is also detected by picturing the algae . The micro algal mapping architecture shows the proper work flow of sensing images of micro algal outbreak the tools such as support vector machine is also used in this process . The Initial Classification Using CNN , Object Detection Integration and Hybrid Decision Fusion Mechanisms are utilized in the process . Almost 90% of Accuracy , Precision and Recall can be acquired by analyzing the Remote Sensing Images . A RESISC method is guided by the object based on the common use of a proposed learning and detector classification