Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal IAES International Journal of Artificial Intelligence (IJ-AI) Indonesian Journal of Electrical Engineering and Computer Science
Sathyanarayana, Natya
Unknown Affiliation
Computer Science & IT Engineering

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Hybrid adaptive neural network for remote sensing image classification Sathyanarayana, Natya; Singh, Seema
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 13, No 2: June 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v13.i2.pp2291-2300

Abstract

The proposed study employed a method for identifying the main contents (category/class) that a remote sensing image (RSI) belongs to, as well as the percentage contribution if the image comprises a significant number of different content types. Histogram based approach has been used to extract the pixel density distribution (PDD) and its normalized form helps to make solution independent from image physical characteristics. A multilayer feedforward artificial neural network (ANN) design has been used to address the classification problem. The architecture included an adaptive form of transfer function, whose slope characteristics changes along with weights as learning progresses. The approach of solution design is computation efficient because it doesn’t apply extensive pre-processing.
Mysore sentinel-2: deep learning for image classification with optimizer exploration Sathyanarayana, Natya; Singh, Seema

Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v34.i1.pp647-657

Abstract

The classification of Sentinel-2 image is presented in this work using a tile based methodology. The Mysore district of India's Karnataka state serves as the subject region of this research. By tiling Sentinel-2 images, we were able to construct a distinct dataset and get approximately 3,000 training samples for the five classes. These images are manually labelled and geo-referenced. Three different optimizers were employed in a thorough analysis with deep learning models such as ResNet50, MobileNetV2, ShuffleNet, and VGG16 to achieve better performance metrics. With a classification accuracy of 98.1% on RESNet50 using ADAM surpassed the others. This facilitates investigating various geographical data analytics applications of the study region.
Co-Authors Singh, Seema