Detection of changes using bi-temporal multispectral images is an important processing tool and widely used to predict changes over the earth’s surface and for various applications in agricultural ...and environmental monitoring, natural hazards assessment, urban drafting, and map redrafting. All the traditional methods used to detect changes are not realistic to predict real changes. Therefore, finding actual changes in satellite images is still a challenging task. In the solution, the real changes occurring on the Earth’s surface are being explored using artificial intelligence techniques. To complete this challenging task, we recommend a novel machine intelligence learning model based on terrestrial prediction for multispectral satellite image change detection. The main objective of our algorithm is to discover the changes without any additional computations, which has the maximum learning characteristics from bi-temporal satellite images. Machine intelligence learning model has been trained with seven transposed features of training data set and used to predict for bi-temporal image target data set and ultimately obtained good quality difference image. The accuracy of the designed machine intelligence learning model is obtained 99.94% with a Kappa coefficient of 0.9985. The water expanded reformed geographical area near Poyang Lake is calculated with 1749.918628 km
2
. Due to the sudden rise of water, there was a lot of damage near Poyang Lake, which can be estimated from the enlarged water area. Such a good accuracy has been obtained from the proposed method, which is much better than other previous methods. This proposed method can be applied to correctly estimate any unexpected damage or change.
A spatial ETL tool that allows interoperability between spatial and non-spatial data is presented in this article. The primary goal of the tool is to provide spatial data processing and ...transformation among various data formats. This is made possible by the ETL process, which extracts, transforms and loads data. The use of spatial data has become significant in everyday life, because only correctly applying the data enables users to extract the true value spatial data offers. The main purpose of this article is to demonstrate the capability and usability of the spatial ETL tool, in order to introduce a more detailed definition of the ETL process to acquaint the reader with the FME Desktop tool, and to demonstrate the applicability of the tool in two case studies. In the first case, a unified spatial data warehouse is built from non-homogeneous data warehouses in order to assess the impacts and effects the geological basis had on the amount of damage to buildings in the 2004 earthquake. The second case demonstrates how the spatial ETL tool can be used to inform locals of predicted spatial changes in the area. The flexibility and the efficiency of the spatial ETL tool are successfully demonstrated in both cases; ETL turns out to be a robust tool for editing and analysing data.