A Survey: Image processing in Real-Time Big Data Analytical Architecture

Remote sensors generate the huge amount of data from satellites. Nowadays, there is a large demand for real-time data for remote sensing applications, and to extract useful information from the data. This paper examines different mining approaches in various satellite image processing applications. A two level merging approach is used to extract the sea area. But, the performance is lower due to the lack of feature extraction. The sensors are deployed to feature out the temperature and pressure for the fire detection. It results in the prediction of fire and is of the high cost. Whereas, Sea-land segmentation using meteorological datasets are carried out only offline. Thus, it is necessary to design an architecture that supports both offline and the real-time data analysis. In addition, scale invariant feature transform and various algorithms can be used as to elaborate the working of real-time data analytical architecture.