Search research articles

Search research articles

Related Experiment Video

Updated: Jan 9, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Context-Aware Deep Learning based Indian Footpath Damage Segmentation Dataset for Risk Assessment.

Priti Chakurkar¹, Deepali Vora²

¹Symbiosis Institute of Technology, Pune Campus, Symbiosis International (Deemed University), Pune, 412115, India.

Scientific Data

|December 8, 2025

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same journal

A dataset of four Decades of Italian TV Commercials: visual, audio, and linguistic feature descriptors.

Scientific data·2026

Same journal

A telomere-to-telomere gap-free genome assembly of the indica rice (Oryza sativa L.) cytoplasmic male sterile line Funong A.

Scientific data·2026

Same journal

A large-scale heterogeneous 3D magnetic resonance brain imaging dataset for self-supervised learning.

Scientific data·2026

Same journal

ECHIDNA: Extreme Climate Historical and Future Indices Data under Numerous Approaches across Major Chinese River Basins Based on CMIP6 Multi-Model Ensemble.

Scientific data·2026

Same journal

An open fMRI dataset examining the effects of online social and non-social information distraction on attention.

Scientific data·2026

Same journal

A comprehensive dataset of 32 million pentapeptide structures for high-throughput virtual screening.

Scientific data·2026

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

This study introduces a new dataset for detecting footpath damage, crucial for preventing pedestrian falls in cities. The dataset, featuring detailed annotations, aids in developing automated infrastructure monitoring systems.

Area of Science:

Computer Vision
Urban Infrastructure Monitoring
Civil Engineering

Background:

Damaged footpaths present significant pedestrian safety risks in urban areas.
Automated monitoring systems are needed for infrastructure maintenance and fall prevention.
Existing datasets may lack the detail or diversity required for robust model training.

Purpose of the Study:

To introduce the Indian Footpath Damage Segmentation Dataset for research in automated infrastructure monitoring.
To provide a high-resolution, manually annotated dataset for footpath damage detection and severity assessment.
To establish a benchmark for evaluating deep learning models in this domain.

Main Methods:

Collected a high-resolution image dataset of footpath damage in Pune, India, under varied conditions.

Related Experiment Videos

Last Updated: Jan 9, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Manually annotated images with pixel-level segmentation masks for footpath damage, including severity levels.

Trained and evaluated deep learning models, including U-Net with EfficientNet-B3 and ensemble methods, on the dataset.

Main Results:

The best-performing ensemble model achieved a Dice score of 0.6899 and an IoU of 0.6741.
The model demonstrated strong performance with an accuracy of 0.9317 and an F1-score of 0.6899.
The dataset facilitates the development of accurate footpath damage segmentation models.

Conclusions:

The Indian Footpath Damage Segmentation Dataset is a valuable resource for advancing research in pedestrian safety and infrastructure monitoring.
The developed models show promising results for automated detection and severity assessment of footpath damage.
The dataset is publicly available on Zenodo to encourage further research and development.