Signature Library Creation from Satellite Imagery for Automatic Target Recognition
- Resource Type
- Conference
- Authors
- Sakshi; Gupta, Charu; Patra, Tushar Kanti; Tomar, Sahil; Singh Tomar, Vivek Kumar
- Source
- 2024 IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI) Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), 2024 IEEE International Conference on. 2:1-6 Mar, 2024
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Engineered Materials, Dielectrics and Plasmas
Fields, Waves and Electromagnetics
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Transportation
Technological innovation
Image resolution
Target recognition
Feature extraction
Libraries
Satellite images
Classification algorithms
ATR
Signature Library
Feature Matching
Template Matching
SIFT
ORB
- Language
The development of efficient algorithms for automatic target recognition (ATR) and classification from satellite imageries are of great challenges and are of particular importance for military applications. These algorithms are crucial for extraction of various types of military targets and their geo-location from varying resolution panchromatic, multispectral satellite images. Traditionally the signature libraries of military targets play a key role in determining the accuracy and performance of ATR algorithms. The signature libraries are comprised of target chips (template) or set of features or templates and features both in a hybridized ATR method. In this work, we have developed a unique method for creating exhaustive sets of signature libraries for variety of military targets. This work demonstrates that using our method of creating signature libraries greatly helps in enhancing the ATR results in a hybridized method and to achieve an accuracy level of over 95% in detection result.