Applications of Hyperspectral Remote Sensing, GIS, and Artificial Intelligence in Agriculture
Mamta J Patange,G J Abhishek,T R Ashwini,Tanu Shree Lakra,Lalta Prasad Verma,Anjali Dutt,Magendra Pal Singh,Kajal Kushwaha,P C Chanyal,Ashwini T R
DOI: https://doi.org/10.9734/acri/2024/v24i7823
2024-08-17
Archives of Current Research International
Abstract:There is a global need for a new approach that can help in solving the problems of food and water shortage, which are significantly affected by population growth and climatic changes. The conventional methods that are used for evaluating and mentoring different agricultural activities and processes have several challenges. These methods are laborious, destructive, time-consuming, and cost-consuming. Therefore, an integration of different approaches, such as hyperspectral remote sensing (HRS), Geographic Information Systems (GIS), and artificial intelligence (AI) has been found to be a very effective tool for enhancing agricultural productivity as well as sustainability. The main objective of this review is to demonstrate the very advanced applications and achievements of these techniques in the field of agricultural activities, as well as their potentialities in precision agriculture (PA). The HRS sensors acquire detailed spectral data, which can be used in several applications, such as crop monitoring and evaluating soil fertility, as well as providing valuable outputs to help in natural resource management. On the other hand, the GIS technique manages the spatial information, is combined with the attributes of the vegetation cover, water bodies, bare soils, etc., and applies statistical and mathematical spatial models for mapping and modeling purposes in order to enable better decision-making for all agricultural practices. Additionally, AI tools, including machine learning (ML) as well as deep learning (DL), are used for the spatial, spectral, wet chemistry, environmental, and field data processing and modeling to find the best model that can be automatically utilized in management solutions. Furthermore, the article demonstrates the limitations, challenges, and future directions of these approaches. Moreover, emphasizing the critical need for interdisciplinary contribution between the researchers, government, and farmers can optimize agricultural outcomes and address environmental concerns. Therefore, an integration of these approaches is considered as a very effective tool for detecting, characterizing, estimating, and mapping several objects using the mapping tools in the environments of different spatial analysis techniques and software. However, utilizing and privileging these techniques provide crucial and essential benefits in order to achieve better environmental resources management and agricultural sustainability.