NORMA eResearch @NCI Library

Effective Image-Based Parking Occupancy Detection using Masked Region Based Convolutional Neural Network

Skariah, Ronu (2022) Effective Image-Based Parking Occupancy Detection using Masked Region Based Convolutional Neural Network. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (1MB) | Preview


As the field of machine learning is shaping the world completely the usage of the techniques in machine learning can be effectively utilized to finding out a solution to the parking space identification problem which is less expensive and easier to implement. This research implements a dynamic, simple, and less expensive algorithm for the identification of parking spaces using machine learning techniques. This algorithm uses a deep learning network for the identification of all the parking spaces in any parking lot and performs the intersection over union technique to identify all the empty and filled space in any parking lot. This dynamic algorithm was able to perform on an average scale over images of different parking lots collected over sunny, rainy, and overcast weather. The object detection deep learning network Mask-RCNN which uses instance segmentation for the identification of vehicles was performed with a map score of 0.901 after performing the hyperparameter tunning-based training over 2,000 images.

Item Type: Thesis (Masters)
Cosgrave, Noel
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Algebra > Algorithms > Computer algorithms
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 26 May 2023 15:14
Last Modified: 26 May 2023 15:14

Actions (login required)

View Item View Item