JJEM: Special Edition 2 (December 2024)
JJEM: Special Edition 2 (December 2024)
2024-12-08
An Innovative Method for Categorizing Satellite Images via Enhanced Deep Convolutional Neural Network.
Sahana S H, Dr. Raghavendra S P
Satellite image categorization plays a optimal role in various scenarios includes natural disaster response, earth monitoring, prevention of hazardous events and analysis of terrestrial biodiversity. CNN algorithms employed on deep learning were exploited to classify orbital images into respective categorize. In addition to this study categorize satellite photos but it can also categorize obvious classes and recognizing the specific traits of those cataloging categories. The main issue with satellite photography is that different satellite images may have different characteristics, which makes space probe image categorization challenging. Another issue is that the majority of safe light images include signal purification over-the-air imagery noise structure are estimated using the CNN mode. The proposed system involves the convolution neural network and it is implemented in python through the tenser flow and keras libraries. In this article carried out ResNet-50 model, which was able to enrich the classification feasibility more than 96% accuracy.
Convolutional neural network, Remote sensing, Image Processing, Deep Learning