Effective Detection of Materials in Construction Drawings using YOLOv4-based Small Object Detection Techniques
Jiwoo Sim, Hee-Jo Woo, YunHwan Kim, Eung-Tae Kim · 2022
Since the quantity surveying of the materials marked on construction drawings is conducted manually, it is very time-consuming and causes problems such as incorrect calculation transactions. So, a fast and accurate AI-based automatic quantity surveying system is required. In order to accurately detect steel materials in construction drawings, we propose data augmentation techniques and spatial attention modules for improving small object detection performance based on YOLOv4. Experimental results show that the proposed method increases the accuracy and precision by 1.8% and 16% respectively compared with the conventional YOLOv4.