A platform for research: civil engineering, architecture and urbanism
Vision-based method for tracking workers by integrating deep learning instance segmentation in off-site construction
Abstract Vision-based tracking of construction workers is the fundamental step for many automated applications in off-site construction. This research proposes a vision-based method for tracking workers in off-site construction by integrating deep learning instance segmentation. The proposed method consists of three main modules including instance segmentation, instance association, and instance assignment. The instance segmentation module applies the Mask R-CNN algorithm to extract masks and bounding boxes of workers from videos. Then, an association matrix is constructed at each two consecutive frames based on the mask intersection-over-union and Kalman filtering. Finally, the instance assignment module solves the association matrix to produce tracking results. In experiments, the proposed method achieved the multiple object tracking accuracy of 96.4% and multiple object tracking precision of 86.2%. The testing results indicate the developed method can successfully track multiple workers when facing the challenges of occlusions and scale variations, etc.
Highlights A vision-based method is proposed for tracking workers in off-site construction. Deep learning instance segmentation and mask association are integrated for tracking. The results achieved at 96.4% on MOTA and 86.2% on MOTP on nine testing videos. The proposed method achieved MOTA of 95.6% and MOTP of 85.3% on dealing with occlusions.
Vision-based method for tracking workers by integrating deep learning instance segmentation in off-site construction
Abstract Vision-based tracking of construction workers is the fundamental step for many automated applications in off-site construction. This research proposes a vision-based method for tracking workers in off-site construction by integrating deep learning instance segmentation. The proposed method consists of three main modules including instance segmentation, instance association, and instance assignment. The instance segmentation module applies the Mask R-CNN algorithm to extract masks and bounding boxes of workers from videos. Then, an association matrix is constructed at each two consecutive frames based on the mask intersection-over-union and Kalman filtering. Finally, the instance assignment module solves the association matrix to produce tracking results. In experiments, the proposed method achieved the multiple object tracking accuracy of 96.4% and multiple object tracking precision of 86.2%. The testing results indicate the developed method can successfully track multiple workers when facing the challenges of occlusions and scale variations, etc.
Highlights A vision-based method is proposed for tracking workers in off-site construction. Deep learning instance segmentation and mask association are integrated for tracking. The results achieved at 96.4% on MOTA and 86.2% on MOTP on nine testing videos. The proposed method achieved MOTA of 95.6% and MOTP of 85.3% on dealing with occlusions.
Vision-based method for tracking workers by integrating deep learning instance segmentation in off-site construction
Xiao, Bo (author) / Xiao, Hairong (author) / Wang, Jingwen (author) / Chen, Yuan (author)
2022-01-22
Article (Journal)
Electronic Resource
English
Construction worker , Deep learning , Instance segmentation , Off-site construction , Vision-based tracking , CNN , convolutional neural network , PPE , personal protective equipment , ID , identification , IoU , intersection-over-union , MOTA , multiple object tracking accuracy , MIoU , mask intersection-over-union , ROI , region of interest
Unsupervised cervical cell instance segmentation method integrating cellular characteristics
Springer Verlag | 2024
|