Scribd
Upload
3 views4 pages

DL

The document outlines a project focused on developing an enhanced deep learning algorithm for pavement crack segmentation and severity assessment, addressing the limitations of manual inspection and existing models. The objectives include analyzing current algorithms and creating a new framework to improve accuracy in segmentation and severity estimation. The methodology involves reviewing existing models, preprocessing datasets, implementing baseline models, and evaluating the new framework against state-of-the-art methods.

Uploaded by

Pranay Mulupuri
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views4 pages

DL

The document outlines a project focused on developing an enhanced deep learning algorithm for pavement crack segmentation and severity assessment, addressing the limitations of manual inspection and existing models. The objectives include analyzing current algorithms and creating a new framework to improve accuracy in segmentation and severity estimation. The methodology involves reviewing existing models, preprocessing datasets, implementing baseline models, and evaluating the new framework against state-of-the-art methods.

Uploaded by

Pranay Mulupuri
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

An Enhanced Deep Learning Algorithm for Pavement

Crack Segmentation and Severity Assessment

Pranay Raj Mulupuri


Roll No.: 252SP019

Under the guidance of


Dr. Shyam Lal

Signal processing and Machine Learning


Department of Electronics and Communication Engineering
National Institute of Technology Karnataka, Surathkal

July 23, 2026

Pranay Raj Mulupuri An Enhanced Deep Learning Algorithm for Pavement Crack Segmentation and Severity 1As
Project Details

Problem Statement
Design and Development of Enhanced Deep Learning Algorithm for
pavement crack segmentation and severity assessment

Motivation
Manual pavement inspection is time-consuming, subjective, and
costly.
Existing deep learning models often struggle with precise crack
localization and severity estimation.
Accurate crack segmentation and severity assessment are essential for
timely maintenance and improved road safety.

Pranay Raj Mulupuri An Enhanced Deep Learning Algorithm for Pavement Crack Segmentation and Severity 2As
Objectives and Methodology
Objectives
1 Implement and perform a comparative analysis of existing deep

learning algorithms for pavement crack segmentation and severity


assessment.
2 Design and develop an enhanced deep learning algorithm to improve
segmentation accuracy and crack severity estimation.

Proposed Methodology
Review existing deep learning models for pavement crack analysis.
Preprocess and prepare benchmark pavement crack datasets.
Implement baseline segmentation models.
Develop the enhanced deep learning framework.
Evaluate using IoU, Dice Score, Precision, Recall, F1-score, and
severity assessment accuracy.
Compare the proposed model with existing state-of-the-art methods.
Pranay Raj Mulupuri An Enhanced Deep Learning Algorithm for Pavement Crack Segmentation and Severity 3As
Thank You

Pranay Raj Mulupuri An Enhanced Deep Learning Algorithm for Pavement Crack Segmentation and Severity 4As

You might also like