Abstract: The project aimed at detecting aircrafts using a Convoluted Neural Network. The network was trained and tested which helped provide more accurate results. The inputs into the system were satellite images. This image was processed by being passed through the four layers of the network. The layers carried out the following tasks: i) Removal of Noise ii) Detection of Object iii) Feature Extraction of the object iv) Matrix Analysis so as to deliver the results. The proposed methodology gave the output in the form of the same satellite image along with the object being marked within squares. The functions written were divided to carry out the tasks of image identification and feature extraction. The Neural Network also needed to be trained prior to the detection process in order to ensure that the images would be detected with accuracy. This method in comparison with most other methods yielded better results.
Keywords: Convolutional Neural Network (CNN).
Title: Aircraft Detection in Satellite Images Using a Convolutional Neural Network
Author: Tanmay Kelkar, Umang Sahastransu, Zishen Thajudheen
International Journal of Computer Science and Information Technology Research
ISSN 2348-1196 (print), ISSN 2348-120X (online)
Research Publish Journals
