Image Scene Understanding - Object Detection in Aerial Images using Convolutional Neural Networks |
Author(s): |
| Prof. Nisha Patil , Sandip University; Shivam Gulve, NMIMS,Mukesh Patel School of Technology Management and Engineering; Niketan Bothe, Sandip University |
Keywords: |
| Object detection in Aerial Images, Convolutional Neural Networks, Region-of-Interest pool, SVM |
Abstract |
|
This paper reviews and analyses the approaches concerning information updating from colored aerial photographs with the aim to transmit a detailed database of symbolic information about the objects detected in the aerial images fed to the system. Detecting objects in aerial images is obstructed/challenged by multiple problems such as variance of objects, undetermined obstructions and cluttered background. This paper, analyses the use of Convolutional Neural Networks and its features from multiple layers to perform robust aerial object detection. An image classification-based approach is used to localize the region-of-interest (ROIs) of multiple aspect ratios and further classify them into positive or negative SVM classifiers. |
Other Details |
|
Paper ID: IJSRDV8I100042 Published in: Volume : 8, Issue : 10 Publication Date: 01/01/2021 Page(s): 194-197 |
Article Preview |
|
|
|
|
