Water quality characteristics can be monitored using remote sensing techniques. Geographic information systems use remote sensing photos as well as the results of imaging analysis as major data sources like GIS and LANDSAT-8. Remote sensing enables the simultaneous viewing and mapping of vast areas of the earth's surface when used in conjunction with field surveys. The use of remote sensing for weather forecasting is common in India. Additionally, it is employed to warn people of impending cyclones. It can be used to investigate issues like eutrophication of large bodies of water, oil spills from oil tankers, desertification, air pollution, land degradation, and deforestation. Our issue is the study of water quality using landsat-8 pictures. Artificial intelligence (AI) approaches such as MLP, SVM, and group method of data management were previously used to forecast the components of water quality. A csv file with the data was selected. The best accuracy was connected to the RBD as the kernel function, according to a review of the SVM's structure. The outcomes showed that its precision is sufficient for practical uses. The ARIMA model had the lowest level of accuracy. Our proposed system uses artificial neural networks, such as the Recurrent Neural Network and ARIMA, to determine the water quality in coastal areas utilizing remote sensing data. To achieve a successful result, we select the algorithms that have multilayer perceptron’s. Additionally, utilizing web technologies and frameworks, we create a website that will provide weekly updates on the water quality in a specific location. Water quality monitoring based on remote sensing imagery mostly use the semiempirical technique. The resolution of the majority of remote sensing imageries processed by the semi-empirical technique, on the other hand, is quite poor. For data collecting, the semiempirical technique necessitates a large number of people. Meanwhile, it has time and space constraints since it cannot analyze every picture from diverse imaging settings adaptively. This research presents a unique technique to monitoring water quality by merging high spatial resolution imageries with hyperspectral imageries (HSI). ARIMA Time series is based on a multispectral high spatial resolution picture. The individuals who live close to those water bodies would become more conscious as a result. The areas that will profit from this project are those that are close to water bodies, and those who live there will also be made aware of the problem of water contamination.
Keywords: Remote Sensing, Satellite Image Classification