Published Paper Details:

Considering the enormous difficulties that the globe is experiencing in providing food for an ever-increasing population, the incorporation of technology into agricultural practices has become an absolute need. Climate-controlled agricultural facilities, such as greenhouses and hydroponic systems, provide methods of food production that are both effective and environmentally friendly. It is essential to have the capacity to monitor and control plant health in order to get optimal results while optimizing these habitats. The purpose of this study is to investigate the use of machine learning techniques for the categorization of plants in controlled agricultural situations with the intention of improving precision agriculture operations. One of the most important components of a plant is water. Along these lines, the development of plants is extremely dependent on the changes that occur in the amount of water that is contained within the plant. There have been a lot of different approaches that have been developed in order to enhance plant development in times of water scarcity and dry period pressure. The purpose of this study is to organize and construct an intelligent framework that will promote plant development in a constrained water environment. This framework will be based on Support Vector Machine (SVM) and machine vision. Shading, morphological, and textural highlights were finally isolated from a large number of photographs of turf grass, wheat, and rice plants that were taken under pressure circumstances that were typical of the dry season. At that point, an SVM and an ANN were utilized in order to supervise the process of information organization by using them. When the SVM was used as the classifier, the results showed that the general arrangement exactness of ANN was 92%, and greater correctness were obtained when the SVM was used. The general precision of the SVM was 98.00% for both fresh and wilted plant conditions.