A Deep Learning-Based Approach to Enable Action Recognition for Construction Equipment
In order to support smart construction, digital twin has been a well-recognized concept for virtually representing the physical facility. It is equally important to recognize human actions and the movement of construction equipment in virtual construction scenes. Compared to the extensive research o...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Advances in Civil Engineering |
| Online Access: | http://dx.doi.org/10.1155/2020/8812928 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832546814139564032 |
|---|---|
| author | Jinyue Zhang Lijun Zi Yuexian Hou Mingen Wang Wenting Jiang Da Deng |
| author_facet | Jinyue Zhang Lijun Zi Yuexian Hou Mingen Wang Wenting Jiang Da Deng |
| author_sort | Jinyue Zhang |
| collection | DOAJ |
| description | In order to support smart construction, digital twin has been a well-recognized concept for virtually representing the physical facility. It is equally important to recognize human actions and the movement of construction equipment in virtual construction scenes. Compared to the extensive research on human action recognition (HAR) that can be applied to identify construction workers, research in the field of construction equipment action recognition (CEAR) is very limited, mainly due to the lack of available datasets with videos showing the actions of construction equipment. The contributions of this research are as follows: (1) the development of a comprehensive video dataset of 2,064 clips with five action types for excavators and dump trucks; (2) a new deep learning-based CEAR approach (known as a simplified temporal convolutional network or STCN) that combines a convolutional neural network (CNN) with long short-term memory (LSTM, an artificial recurrent neural network), where CNN is used to extract image features and LSTM is used to extract temporal features from video frame sequences; and (3) the comparison between this proposed new approach and a similar CEAR method and two of the best-performing HAR approaches, namely, three-dimensional (3D) convolutional networks (ConvNets) and two-stream ConvNets, to evaluate the performance of STCN and investigate the possibility of directly transferring HAR approaches to the field of CEAR. |
| format | Article |
| id | doaj-art-7e47eee23f8e4d35a5e85bb7299bedce |
| institution | Kabale University |
| issn | 1687-8086 1687-8094 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advances in Civil Engineering |
| spelling | doaj-art-7e47eee23f8e4d35a5e85bb7299bedce2025-02-03T06:46:57ZengWileyAdvances in Civil Engineering1687-80861687-80942020-01-01202010.1155/2020/88129288812928A Deep Learning-Based Approach to Enable Action Recognition for Construction EquipmentJinyue Zhang0Lijun Zi1Yuexian Hou2Mingen Wang3Wenting Jiang4Da Deng5College of Management and Economics, Tianjin University, Tianjin, ChinaGuangzhou Metro Design and Research Institute Co., Ltd., Guangzhou, ChinaCollege of Intelligence and Computing, Tianjin University, Tianjin, ChinaCollege of Intelligence and Computing, Tianjin University, Tianjin, ChinaGuangzhou Metro Design and Research Institute Co., Ltd., Guangzhou, ChinaCollege of Intelligence and Computing, Tianjin University, Tianjin, ChinaIn order to support smart construction, digital twin has been a well-recognized concept for virtually representing the physical facility. It is equally important to recognize human actions and the movement of construction equipment in virtual construction scenes. Compared to the extensive research on human action recognition (HAR) that can be applied to identify construction workers, research in the field of construction equipment action recognition (CEAR) is very limited, mainly due to the lack of available datasets with videos showing the actions of construction equipment. The contributions of this research are as follows: (1) the development of a comprehensive video dataset of 2,064 clips with five action types for excavators and dump trucks; (2) a new deep learning-based CEAR approach (known as a simplified temporal convolutional network or STCN) that combines a convolutional neural network (CNN) with long short-term memory (LSTM, an artificial recurrent neural network), where CNN is used to extract image features and LSTM is used to extract temporal features from video frame sequences; and (3) the comparison between this proposed new approach and a similar CEAR method and two of the best-performing HAR approaches, namely, three-dimensional (3D) convolutional networks (ConvNets) and two-stream ConvNets, to evaluate the performance of STCN and investigate the possibility of directly transferring HAR approaches to the field of CEAR.http://dx.doi.org/10.1155/2020/8812928 |
| spellingShingle | Jinyue Zhang Lijun Zi Yuexian Hou Mingen Wang Wenting Jiang Da Deng A Deep Learning-Based Approach to Enable Action Recognition for Construction Equipment Advances in Civil Engineering |
| title | A Deep Learning-Based Approach to Enable Action Recognition for Construction Equipment |
| title_full | A Deep Learning-Based Approach to Enable Action Recognition for Construction Equipment |
| title_fullStr | A Deep Learning-Based Approach to Enable Action Recognition for Construction Equipment |
| title_full_unstemmed | A Deep Learning-Based Approach to Enable Action Recognition for Construction Equipment |
| title_short | A Deep Learning-Based Approach to Enable Action Recognition for Construction Equipment |
| title_sort | deep learning based approach to enable action recognition for construction equipment |
| url | http://dx.doi.org/10.1155/2020/8812928 |
| work_keys_str_mv | AT jinyuezhang adeeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT lijunzi adeeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT yuexianhou adeeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT mingenwang adeeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT wentingjiang adeeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT dadeng adeeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT jinyuezhang deeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT lijunzi deeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT yuexianhou deeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT mingenwang deeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT wentingjiang deeplearningbasedapproachtoenableactionrecognitionforconstructionequipment AT dadeng deeplearningbasedapproachtoenableactionrecognitionforconstructionequipment |