ViSnow: Snow-covered Urban Roads Dataset for Computer Vision Applications

Citation Author(s):
Mohamed
Karaa
Systems Engineering Department, Ecole de Technologie Supérieure (ÉTS), Montreal, Canada
Hakim
Ghazzai
King Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia
Lokman
Sboui
Systems Engineering Department, Ecole de Technologie Supérieure (ÉTS), Montreal, Canada
Submitted by:
mohamed karaa
Last updated:
Wed, 04/17/2024 - 10:40
DOI:
10.21227/9dyz-x716
Data Format:
Research Article Link:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

We present ViSnow: a large image dataset for snow-covered roads in an urban setting. The dataset includes an extensive collection of images from traffic surveillance cameras installed in Montreal, Quebec, Canada, during the winters of 2022 and 2023. ViSnow dataset aims to enable computer vision applications in intelligent transportation and winter road maintenance. ViSnow comprises 294,000 images describing various settings spanning day and night periods, different weather conditions (snow, rain, clear), and multiple urban areas (residential, commercial, industrial). We attach a metadata JSON file recording the timestamp and weather data to each image to provide more contextual information. ViSnow images are annotated to describe four snow cover classes: “clear surface”, “light-covered surface”, “medium-to-heavy-covered surface”, and “plowed surface”, matching possible snow removal operations.

Instructions: 

For citation and instructions to use the dataset please refer to:

M. Karaa, H. Ghazzai, and L. Sboui, "ViSnow: Snow-covered Urban Roads Dataset for Computer Vision Applications", IEEE Open Journal of Systems Engineering, Apr. 2024.

M. Karaa, H. Ghazzai, and L. Sboui, "A Dataset Annotation System for Snowy Weather Road Surface Classification", IEEE Open Journal of Systems Engineering, Apr. 2024.