Show simple item record

dc.contributor.authorNabirye, Sharifah
dc.date.accessioned2023-08-16T08:29:45Z
dc.date.available2023-08-16T08:29:45Z
dc.date.issued2023-07-07
dc.identifier.citationNabirye, Sharifah. (2023). Development of a deep learning model for detection of road accidents. (Unpublished undergraduate dissertation) Makerere University; Kampala, Uganda.en_US
dc.identifier.urihttp://hdl.handle.net/20.500.12281/16222
dc.descriptionA research report submitted to the College of Engineering Design and Art in partial fulfillment of the requirement for the award of the degree Bachelor of Science Electrical Engineering of Makerere University.en_US
dc.description.abstractThis project aims to develop a deep learning model for the detection of road accidents. The study focuses on leveraging the power of deep learning techniques to improve the accuracy and efficiency of accident detection systems, ultimately contributing to enhanced road safety. The methodology employed in this project involves the collection of a diverse dataset consisting of images and videos captured from various traffic scenarios. The dataset is annotated to label accident-related instances, enabling the training of a deep learning model. Convolutional Neural Networks (CNNs) are utilized to extract relevant features from the input data and identify patterns indicative of road accidents. Transfer learning is applied to leverage pre-trained models and optimize the training process. The developed model is deployed as a web application, allowing users to access the accident detection system conveniently through web browsers. Furthermore, a reporting system using SMS (Short Message Service) is incorporated to enable automatic alerts and notifications to emergency services and relevant stakeholders in real-time. The key results demonstrate the effectiveness of the developed deep learning model in detecting road accidents. Through rigorous evaluation on a large-scale dataset, the model achieves a high detection accuracy and exhibits robust performance across different road and weather conditions. The model's efficiency is also highlighted, as it provides real-time accident detection capabilities, enabling prompt responses from emergency services. Based on the findings, the study concludes that deep learning models can significantly contribute to the detection of road accidents. By accurately identifying accidents, emergency services can be alerted promptly, potentially reducing response times and improving overall road safety. Furthermore, this project recommends the integration of the developed deep learning model into existing traffic management systems. By deploying this model across various road networks, authorities can enhance their accident detection capabilities and provide better support to accident victimen_US
dc.language.isoenen_US
dc.publisherMakerere Universityen_US
dc.subjectDeep learning modelen_US
dc.subjectRoad accidentsen_US
dc.titleDevelopment of a deep learning model for detection of road accidentsen_US
dc.typeThesisen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record