Loss function computation using machine learning algorithms based on the effects of natural disasters and plant diseases on plant growth

Abstract

The perfect place for the human beings to live is the planet named Earth.For this, Earth has to have all the things in the perfect balance for humans or any other living thing to live.Among the things required, plant is the most vital required part.It is the single most important source of oxygen for the living things including humans, animals.On the other hand, Natural Disasters(particular weather factors) are happening frequently and causing varying level of damage to plants.Furthermore, various diseases also harm these plants.For tackling these,here using Machine Learning algorithms, we proposed a “Loss Function” which provides a Loss factor value between (0-1) for determining how certain factors affect plant structure mainly growth and also tell us how much of the Plant is affected/damaged.Many Machine Learning( ML) architectures have been in use for detecting soil structure, plant diseases and other plant related tasks for many years.For the thesis, our group is going to study on how different Natural Disasters (particular weather factors) and Plant Diseases affect plant structure (growth) and get an output of a Loss factor value from the function parameters. Other parameters such as Market demand of plant, economic status of person will also be considered for future work. The Loss function mainly provides a value which is the Loss factor to determine plant growth in a particular condition. Moreover, for the Disease detection we are going to use images with real life backgrounds which will ensure that there are plants in a particular background and we can still detect the disease. Therefore, our main target will be making the “Loss Function” depending on two factors namely Natural Disasters (particular weather factors) and Plant Diseases. By using our “Loss Function” a Loss factor value will be given as output which considers effects of these two parameters on plant growth.