Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 11 (2022) > Article

DOI: 10.14704/NQ.2022.20.11.NQ66314

Measuring the accuracy of Satellite Image Dataset using Fusion based Classification with Neural Networks

Swathika R, Radha N


Image classification is a major technique for handling satellite images and their features. Image fusion is a method to handle images, structured information, segmentation and semantic information. Here we used image fusion-based classification techniques for processing images, extracting features and classifying regions. In this work two stream convolution neural network model is proposed for image classification and decision making. It is a multi-objective process to extract visual and textual features. The deep belief entropy is measured in each pixel values by considering spatial resolution in different bands. The experiments are done using Deep learning embedded methods and visual features are extracted. It is a joint feature combination of fusion and classification techniques to experiment satellite images. The confusion matrix is generated for measuring accuracy index. With this proposed joint deep learning approach, the classification accuracy is achieved as 96% on large scale fused satellite image dataset.


Image classification, Fusion, Deep Learning, CNN, Decision Making

Full Text