Paper Details

Name: Gautam Kumar, Piyush Kumar, Lovish Bansal & Shivam Aanand
Country: India

Year: 2026
Volume: Volume-13, Issue-1 (January-June)
Page Number: 394-400
DOI: https://doi.org/10.5281/zenodo.19017914

ABSTRACT—
The field of multimedia applications and processing is very promising and plays an essential role in many type of artificial intelligence applications like in video summarization, image retrieval, image classification, etc. Multimedia methods using Convolutional neural networks, convinced researchers to build systems that can classify images using data without human input. For feature extraction based on deep learning with convolutional neural networks using pre-trained models such as Alexnet, we renew the measures for satellite image classification. On all three datasets, SAT4, SAT6, and UC, the Resnet50 model outperforms the other models. Merced Land. As depicted in Figure 5, after applying the developed model, the accuracy of UC Merced Land dataset achieves 98%, SAT4 achieves 95.8%, and SAT6 achieves 94.1%.

Keywords—Deep Learning method, Image Compression, Autoencoders, (C_N_N), (G_A_N), (P_S_N_R).