施工现场图像分割数据

该数据集专为建筑施工现场的目标检测和实例分割任务设计，包含 21 类施工要素，包括设备、材料等。数据来源于超过 200 个不同施工项目，由专业施工单位通过近景相机和全景相机，获取的影像覆盖各类施工现场的实体。每类对象的实例数量均超过 1000 个，数据集共计近 2 万张影像。该数据集适用于建筑领域的智能监控、施工安全分析等任务，能够帮助模型在实际施工环境中识别并检测多种施工要素。

This dataset is specifically developed for object detection and instance segmentation tasks on construction sites. It encompasses 21 categories of construction elements, such as equipment and materials. The dataset is sourced from more than 200 distinct construction projects, with images collected by professional construction teams using close-range cameras and panoramic cameras. These images cover various physical entities across different construction sites. Each category contains over 1,000 instances, and the total number of images in the dataset reaches nearly 20,000. This dataset is applicable to tasks like intelligent monitoring and construction safety analysis within the construction field, enabling models to identify and detect various construction elements in actual construction environments.

搜集汇总

背景与挑战

背景概述

该数据集专注于建筑施工现场的图像分割，包含21类施工要素（如设备和材料），数据来源于超过200个施工项目，通过近景和全景相机采集，总计近2万张影像，每类对象实例均超过1000个。它适用于目标检测和实例分割任务，支持建筑领域的智能监控和施工安全分析，数据规模为18GB，每半年更新一次。