Automated Visual Recognition of Dump Trucks in Construction Videos

Azar, Ehsan Rezazadeh; McCabe, Brenda

doi:10.1061/(asce)cp.1943-5487.0000179

Cited by 103 publications

(16 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This topic is using computer vision to capture and solve construction problems. Several directions appeared, including the automatic object recognition [46][47][48][49][50][51][52], automated vision tracking [18,[53][54][55][56], vision-based action recognition of specific construction equipment or workers [16,[57][58][59][60], vision interpretation [61][62][63], and using computer vision for specific construction management purposes, such as construction safety and health monitoring [64,65], progress monitoring [17], and performance monitoring [66]. Also, some sensor techniques were used for monitoring workers' activities [67] or construction site status [68].…”

Section: Topic 3: Computer Visionmentioning

confidence: 99%

Bibliometric review of visual computing in the construction industry

Wang

Lin

2020

Vis. Comput. Ind. Biomed. Art

View full text Add to dashboard Cite

In the construction area, visuals such as drawings, photos, videos, and 3D models, play a significant role in the design, build and maintenance of a facility, bringing efficiency to generate, transfer, and store information. Advanced visual computing techniques facilitate the understanding of design contents, work plans, and other types of information shared in the construction industry. Automatic visual data collection and analysis provide many possibilities to the construction industry and a large number of works have investigated how visual computing can improve construction management processes and other problems in the construction area. However, a comprehensive literature review is needed. This study uses bibliometric approaches to review the works published to date, and analyses the development of knowledge, significant research results, and trends. The purpose of this study is to help newcomers to this research field understand knowledge structure and formulate research directions, thereby enhancing knowledge development. From this study, it can be concluded that computer vision is a key axis of improvement. Moreover, building information modeling, laser scanning, and other visualizationrelated techniques are also important in advancing the construction area.

show abstract

Section: Topic 3: Computer Visionmentioning

confidence: 99%

Bibliometric review of visual computing in the construction industry

Wang

Lin

2020

Vis. Comput. Ind. Biomed. Art

View full text Add to dashboard Cite

show abstract

“…Apart from vehicle classification or detection, CNN is also able to recognize the working or idle state of earthwork machines, like excavators or trucks [45], and CNN-TL can benefit earthmoving operations or related construction management [27,46]. Other non-CNN machine vision methods also have applications in related areas, like vehicle collision prediction or construction machine detection, [47][48][49][50]. It can be seen that current vision-based deep learning researches mainly focus on the vehicle classification or state identification of earthwork machinery, and CNN and CNN-TL are widely applied.…”

Section: Vison-based Deep Learning In Related Areasmentioning

confidence: 99%

Deep Learning Model Comparison for Vision-Based Classification of Full/Empty-Load Trucks in Earthmoving Operations

et al. 2019

View full text Add to dashboard Cite

Earthmoving is an integral civil engineering operation of significance, and tracking its productivity requires the statistics of loads moved by dump trucks. Since current truck loads’ statistics methods are laborious, costly, and limited in application, this paper presents the framework of a novel, automated, non-contact field earthmoving quantity statistics (FEQS) for projects with large earthmoving demands that use uniform and uncovered trucks. The proposed FEQS framework utilizes field surveillance systems and adopts vision-based deep learning for full/empty-load truck classification as the core work. Since convolutional neural network (CNN) and its transfer learning (TL) forms are popular vision-based deep learning models and numerous in type, a comparison study is conducted to test the framework’s core work feasibility and evaluate the performance of different deep learning models in implementation. The comparison study involved 12 CNN or CNN-TL models in full/empty-load truck classification, and the results revealed that while several provided satisfactory performance, the VGG16-FineTune provided the optimal performance. This proved the core work feasibility of the proposed FEQS framework. Further discussion provides model choice suggestions that CNN-TL models are more feasible than CNN prototypes, and models that adopt different TL methods have advantages in either working accuracy or speed for different tasks.

show abstract

“…Detection algorithms, usually relying on machine learning techniques, involve training to learn the unique signature of a given object. Algorithms include neural networks (Rezazadeh Azar and McCabe, 2012a), support vector machines with specific feature models (Memarzadeh et al, 2012), random forests (Park et al, 2011a), and parts-based models (Rezazadeh Azar and McCabe, 2012b). Parts-based modeling approaches work best for articulated objects because their appearance geometry has high variation, which can be compensated through multiple, individual part detectors.…”

Section: Detectionmentioning

confidence: 99%

“…Foreground estimation works well when the entrance into the scene of the object is controlled, or the object of interest is not occluded by other foreground objects (Yang et al, 2011). In less controlled settings, stronger results are obtained by combining the two techniques (Gong and Caldas, 2011;Chi and Caldas, 2011;Rezazadeh Azar and McCabe, 2012a). The value is due to the fact that foreground detection methods do not classify the detected objects.…”

Section: Detectionmentioning

confidence: 99%

Fusion of Photogrammetry and Video Analysis for Productivity Assessment of Earthwork Processes

Bügler

Borrmann

Ogunmakin

et al. 2016

Computer aided Civil Eng

View full text Add to dashboard Cite

The high complexity of modern large‐scale construction projects leads their schedules to be sensitive to delays. At underground construction sites, the earthwork processes are vital, as most of the following tasks depend on them. This article presents a method for estimating the productivity of soil removal by combining two technologies based on computer vision: photogrammetry and video analysis. Photogrammetry is applied to create a time series of point clouds throughout excavation, which are used to measure the volume of the excavated soil for daily estimates of productivity. Video analysis is used to generate statistics regarding the construction activities for estimating productivity at finer time scales, when combined with the output from the photogrammetry pipeline. As there may be multiple causes for specific productivity levels, the automated generation of progress and activity statistics from both measurement methods supports interpretation of the productivity estimates. Comparison to annotated ground truth for the tracking and activity monitoring method highlights the reliability of the extracted information. The suitability of the approach is demonstrated by two case studies of real‐world urban excavation projects.

show abstract

Automated Visual Recognition of Dump Trucks in Construction Videos

Cited by 103 publications

References 32 publications

Bibliometric review of visual computing in the construction industry

Bibliometric review of visual computing in the construction industry

Deep Learning Model Comparison for Vision-Based Classification of Full/Empty-Load Trucks in Earthmoving Operations

Fusion of Photogrammetry and Video Analysis for Productivity Assessment of Earthwork Processes

Contact Info

Product

Resources

About