Poplar, Grandis Cross-Cut (Grandis-CC), and Grandis Radial-Cut (Grandis-RC)
收藏资源简介:
该研究构建了包含杨木、大叶桉横切与大叶桉径切的三类木材多模态数据集,由丹麦技术大学等机构通过精密实验采集。数据集规模涵盖数百个样本,每个样本包含RGB图像与对应热响应图像,空间分辨率分别为176×176与128×128像素,并辅以温控测试台温度图。数据采集过程采用标准化协议,通过铝制温控平台、RGB相机与红外相机同步记录,经加权平衡与帧平均处理获得准稳态热响应。该数据集旨在支持物理信息深度学习框架的开发,解决木材空间异质性热响应的像素级预测问题,推动木材热性能分析与自适应建筑设计。
This study develops a multimodal dataset covering three wood types: poplar, transverse-cut *Eucalyptus robusta*, and radial-cut *Eucalyptus robusta*. The dataset was collected via precise experiments by institutions including the Technical University of Denmark and other relevant organizations. It contains hundreds of samples, with each sample comprising RGB images and corresponding thermal response images, which have spatial resolutions of 176×176 and 128×128 pixels respectively, alongside temperature maps acquired from the temperature-controlled test bench. The data collection follows a standardized protocol, where synchronized recording is conducted using an aluminum temperature-controlled platform, RGB camera, and infrared camera. Quasi-steady-state thermal responses are subsequently obtained through weighted balancing and frame averaging processing. This dataset aims to support the development of physics-informed deep learning frameworks, address the pixel-level prediction issue of thermal responses for wood spatial heterogeneity, and promote wood thermal performance analysis and adaptive architectural design.




