TY - JOUR ID - TI - Adaptive Hyper Classification Technique for Satellite Images AU - Hazeem B. Taher AU - Najlaa M. Mohie PY - 2019 VL - 9 IS - 2 SP - 32 EP - 41 JO - Journal of Education for Pure Science مجلة التربية للعلوم الصرفة SN - 20736592 AB - Abstract:The satellite image classification system intends to differentiate between the objects being present in theimage. It is highly challenging because the coverage area of the satellite is large such that the objects appear so small.This makes the process of object differentiation complex. Additionally, the classification accuracy is an importantfactor, which the classification system must pass through. This work presents a satellite image classification systemwhich can classify between the vegetation, soil and water bodies, etc. Current research work is in the developmentof a hybrid classification technique for satellite images. High-resolution satellite images that we deal with usingSVM and low accuracy through the use of k-means. It is necessary to provide accuracy and speed of satellite imagesand obtain this through the modulation method suggested through the integration of these technicians, wheretechnology can classify all satellite images .We note that the proposed system is able to classify any satellite imageby highlighting each region and what indicates that area and what percentage when applying the proposed system ofclassification whether the satellite images are based on high accuracy or low accuracy, where the system also showsthe time taken to classify each satellite image. We note that the time does not exceed 30 seconds for the number ofimages used within our database

ER -