A Timely Object Recognition Method for Construction using the Mask R-CNN Architecture

Shamsollahi, Dena; Moselhi, Osama; Khorasani, Khashayar

doi:10.22260/isarc2021/0052

Cited by 5 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Mask-RCNN model was employed in [13] to detect and monitor the progress of work in the construction industry. They used a custom dataset of 1143 images.…”

Section: Related Studiesmentioning

confidence: 99%

See 1 more Smart Citation

Enhancing Object Detection for VIPs Using YOLOv4_Resnet101 and Text-to-Speech Conversion Model

Alahmadi

Rahman

Alkahtani

et al. 2023

MTI

View full text Add to dashboard Cite

Vision impairment affects an individual’s quality of life, posing challenges for visually impaired people (VIPs) in various aspects such as object recognition and daily tasks. Previous research has focused on developing visual navigation systems to assist VIPs, but there is a need for further improvements in accuracy, speed, and inclusion of a wider range of object categories that may obstruct VIPs’ daily lives. This study presents a modified version of YOLOv4_Resnet101 as backbone networks trained on multiple object classes to assist VIPs in navigating their surroundings. In comparison to the Darknet, with a backbone utilized in YOLOv4, the ResNet-101 backbone in YOLOv4_Resnet101 offers a deeper and more powerful feature extraction network. The ResNet-101’s greater capacity enables better representation of complex visual patterns, which increases the accuracy of object detection. The proposed model is validated using the Microsoft Common Objects in Context (MS COCO) dataset. Image pre-processing techniques are employed to enhance the training process, and manual annotation ensures accurate labeling of all images. The module incorporates text-to-speech conversion, providing VIPs with auditory information to assist in obstacle recognition. The model achieves an accuracy of 96.34% on the test images obtained from the dataset after 4000 iterations of training, with a loss error rate of 0.073%.

show abstract

“…The Mask-RCNN model was employed in [13] to detect and monitor the progress of work in the construction industry. They used a custom dataset of 1143 images.…”

Section: Related Studiesmentioning

confidence: 99%

“…The size of input images provided for training [ (10,13), (16,30) Learning rate lr_decay_epoch = 50 Every 50 epochs, the learning rate decays batch_size = 32, 64…”

Section: Yolov4_resnet101mentioning

confidence: 99%

Enhancing Object Detection for VIPs Using YOLOv4_Resnet101 and Text-to-Speech Conversion Model

Alahmadi

Rahman

Alkahtani

et al. 2023

MTI

View full text Add to dashboard Cite

show abstract

“…However, these models largely focus on construction works that can be viewed externally and studies incorporating indoor progress monitoring are limited. In recent CV-based indoor construction progress monitoring studies, Mask R-CNN models have been applied to recognise the building objects (walls, doors and lifts) (Ying and Lee, 2019) and HVAC ducts (Shamsollahi et al , 2021). Mask R-CNN has also been used for calculating the work-in-progress of brick layering and plastering of an indoor wall (Wei et al , 2022).…”

Section: Literature Reviewmentioning

confidence: 99%

“…The pioneering studies of Ying and Lee (2019) and Shamsollahi et al (2021) only provide evidence on the recognition of the indoor elements using Mask R-CNN. The automation level is heavily manually intervened during the data collection in Ying and Lee (2019).…”

Section: Comparison With the Previous Studies Which Used DL Modelsmentioning

confidence: 99%

“…The automation level is heavily manually intervened during the data collection in Ying and Lee (2019). As synthetic images are used as the training images, Shamsollahi et al (2021) fails to reflect the impacts of challenging indoor construction environment. While Wei et al (2022) calculates indoor work completion percentage using Mask R-CNN-based segmentation and maps the relationship between two-dimensional images and three-dimensional building information models, the high manual intervention in the data collection and algorithmic pre-processing Compared to these studies, the automation level of the current study is high due to the use of fixed time-lapse cameras for data collection and zero manual intervention from the pre-processing algorithms to enhance visual quality of indoor images.…”

Section: Comparison With the Previous Studies Which Used DL Modelsmentioning

confidence: 99%

See 1 more Smart Citation

A deep learning-based approach to facilitate the as-built state recognition of indoor construction works

Ekanayake

Fini

Wong

et al. 2022

View full text Add to dashboard Cite

Purpose Recognising the as-built state of construction elements is crucial for construction progress monitoring. Construction scholars have used computer vision-based algorithms to automate this process. Robust object recognition from indoor site images has been inhibited by technical challenges related to indoor objects, lighting conditions and camera positioning. Compared with traditional machine learning algorithms, one-stage detector deep learning (DL) algorithms can prioritise the inference speed, enable real-time accurate object detection and classification. This study aims to present a DL-based approach to facilitate the as-built state recognition of indoor construction works. Design/methodology/approach The one-stage DL-based approach was built upon YOLO version 4 (YOLOv4) algorithm using transfer learning with few hyperparameters customised and trained in the Google Colab virtual machine. The process of framing, insulation and drywall installation of indoor partitions was selected as the as-built scenario. For training, images were captured from two indoor sites with publicly available online images. Findings The DL model reported a best-trained weight with a mean average precision of 92% and an average loss of 0.83. Compared to previous studies, the automation level of this study is high due to the use of fixed time-lapse cameras for data collection and zero manual intervention from the pre-processing algorithms to enhance visual quality of indoor images. Originality/value This study extends the application of DL models for recognising as-built state of indoor construction works upon providing training images. Presenting a workflow on training DL models in a virtual machine platform by reducing the computational complexities associated with DL models is also materialised.

show abstract

Integrated Framework Using Computer Vision and Ultra-Wide Band Techniques for Progress Reporting in Construction Projects

Shamsollahi,

Moselhi,

Khorasani

2024

Lecture Notes in Civil Engineering

View full text Add to dashboard Cite

A Timely Object Recognition Method for Construction using the Mask R-CNN Architecture

Cited by 5 publications

References 0 publications

Enhancing Object Detection for VIPs Using YOLOv4_Resnet101 and Text-to-Speech Conversion Model

Enhancing Object Detection for VIPs Using YOLOv4_Resnet101 and Text-to-Speech Conversion Model

A deep learning-based approach to facilitate the as-built state recognition of indoor construction works

Integrated Framework Using Computer Vision and Ultra-Wide Band Techniques for Progress Reporting in Construction Projects

Contact Info

Product

Resources

About