TY  - Data
T1  - A dataset of blowing snow event visibility classification for the Makitasi Wind Zone, China, 2024–2025
A1  - ma lei
A1  - hu zhi xuan
A1  - Zhao Jing
A1  - li peng bo
A1  - zhao qin
A1  - pan xing yu
A1  - li tai zhi
DO  - 10.12072/ncdc.snow.db7564.2026
PY  - 2026
DA  - 2026-07-02
PB  - National Cryosphere Desert Data Center
AB  - Low visibility caused by wind-blown snow disasters is a key factor affecting high-grade highway traffic safety and operating efficiency. Refined visibility classification data is of great significance for model training, management and control strategy formulation, and roadside edge computing equipment deployment. This dataset is based on continuous videos from 4 roadside fixed surveillance cameras at the K264+700-K272+700 section of the G3015 Keta Expressway in Mayitas Wind Area, Xinjiang from November 2024 to March 2025. Video time screening is carried out according to the observation data of the visibility meter, frame extraction is carried out at a frequency of 1 minute/frame, and is prepared using a manual hierarchical labeling method. The dataset consists of a compressed package, containing 8 folders, corresponding to 8 visibility levels (0-100m, 100-200m, 200-300m, 300-400m, 400-500m, 500-1000m, 1000-2000m, NULL). A total of 14679 roadside cameras captured images, with an image spatial resolution of 640×480 pixels. This dataset can be used to characterize the temporal and spatial distribution and change laws of highway visibility under wind-blown snow conditions, and serve traffic meteorological warning and highway dynamic management and control in wind-blown snow-prone areas.
DB  - NCDC
UR  - http://www.ncdc.ac.cn/portal/metadata/0b7968f3-3dad-4ddc-8448-48843d675c3b
ER  -