Development of ergonomic posture recognition technique based on 2D ordinary camera for construction hazard prevention through view-invariant features in 2D skeleton motion

Abstract Outdoor tasks operated by construction workers are physically demanding, requiring awkward postures leading to pain, injury, accident, or permanent disability. Ergonomic posture recognition (EPR) technique could be a novel solution for ergonomic hazard monitoring and assessment, yet non-intrusiveness and applicability in complex outdoor environment are always critical considerations for device selection in construction site. Thus, we choose RGB camera to capture skeleton motions, which is non-intrusive for workers compared with wearable sensors. It is also stable and widely used in an outdoor construction site considering various light conditions and complex working areas. This study aims to develop an ergonomic posture recognition technique based on 2D ordinary camera for construction hazard prevention through view-invariant features in 2D skeleton motion. Based on captured 2D skeleton motion samples in the test-run, view-invariant features as classifier inputs were extracted to ensure the learned classifier not sensitive to various camera viewpoints and distances to a worker. Three posture classifiers regarding human back, arms, and legs were employed to ensure three postures to be recognized simultaneously in one video frame. The average accuracies of three classifiers in 5-fold cross validation were as high as 95.0%, 96.5%, and 97.6%, respectively, and the overall accuracies tested by three new activities regarding ergonomic assessment scores captured from different camera heights and viewpoints were 89.2%, 88.3%, and 87.6%, respectively. The developed EPR-aided construction accident auto-prevention technique demonstrated robust accuracy to support on-site postural ergonomic assessment for construction workers’ safety and health assurance.

[1]  O Karhu,et al.  Correcting working postures in industry: A practical method for analysis. , 1977, Applied ergonomics.

[2]  L McAtamney,et al.  RULA: a survey method for the investigation of work-related upper limb disorders. , 1993, Applied ergonomics.

[3]  Hassan Foroosh,et al.  View-Invariant Action Recognition from Point Triplets , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Wen Yi,et al.  Health Profile of Construction Workers in Hong Kong , 2016, International journal of environmental research and public health.

[5]  Mubarak Shah,et al.  View-Invariant Representation and Recognition of Actions , 2002, International Journal of Computer Vision.

[6]  Heng Li,et al.  Biomechanical analysis of risk factors for work-related musculoskeletal disorders during repetitive lifting task in construction workers , 2017 .

[7]  Rama Chellappa,et al.  View Invariance for Human Action Recognition , 2005, International Journal of Computer Vision.

[8]  Thomas Kraus,et al.  The incremental effect of psychosocial workplace factors on the development of neck and shoulder disorders: a systematic review of longitudinal studies , 2013, International Archives of Occupational and Environmental Health.

[9]  I Kant,et al.  Observations of working postures in garages using the Ovako Working posture Analysing System (OWAS) and consequent workload reduction recommendations. , 1990, Ergonomics.

[10]  Ze-Nian Li,et al.  Successive Convex Matching for Action Detection , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[11]  Ying Wu,et al.  View-independent recognition of hand postures , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[12]  Leon Straker,et al.  In vivo laboratory validation of the physiometer: a measurement system for long-term recording of posture and movements in the workplace , 2010, Ergonomics.

[13]  Alan H. S. Chan,et al.  The Ageing Construction Workforce in Hong Kong: A Review , 2015, IMECS 2015.

[14]  F. G. Evans,et al.  Anatomical Data for Analyzing Human Motion , 1983 .

[15]  Jitendra Malik,et al.  Recognizing action at a distance , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[16]  Luca Iocchi,et al.  Human Posture Tracking and Classification through Stereo Vision and 3D Model Matching , 2008, EURASIP J. Image Video Process..

[17]  Ronald Poppe,et al.  A survey on vision-based human action recognition , 2010, Image Vis. Comput..

[18]  Heng Li,et al.  Identification of Biomechanical Risk Factors for the Development of Lower-Back Disorders during Manual Rebar Tying , 2017 .

[19]  Jake K. Aggarwal,et al.  View invariant human action recognition using histograms of 3D joints , 2012, 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[20]  H Laitinen,et al.  The validity of the TR safety observation method on building construction. , 1999, Accident; analysis and prevention.

[21]  Jun Qiu,et al.  Construction worker's awkward posture recognition through supervised motion tensor decomposition , 2017 .

[22]  Yantao Yu,et al.  Visualization technology-based construction safety management: A review , 2017 .

[23]  Feniosky Peña-Mora,et al.  Vision-Based Detection of Unsafe Actions of a Construction Worker: Case Study of Ladder Climbing , 2013, J. Comput. Civ. Eng..

[24]  J.B. Burns,et al.  View Variation of Point-Set and Line-Segment Features , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[25]  Mubarak Shah,et al.  Human Action Recognition in Videos Using Kinematic Features and Multiple Instance Learning , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  Fei Dai,et al.  Risk Assessment of Work-Related Musculoskeletal Disorders in Construction: State-of-the-Art Review , 2015 .

[27]  Feniosky Peña-Mora,et al.  Comparative Study of Motion Features for Similarity-Based Modeling and Classification of Unsafe Actions in Construction , 2014, J. Comput. Civ. Eng..

[28]  J S Silva,et al.  Evaluation of lumbar overload in hotel maids. , 2012, Work.

[29]  Mani Golparvar-Fard,et al.  Automated Vision-based Recognition of Construction Worker Actions for Building Interior Construction , 2012 .

[30]  Hong Zhang,et al.  Wearable IMU-based real-time motion warning system for construction workers' musculoskeletal disorders prevention , 2017 .

[31]  Le Thi Hoai An,et al.  Feature selection in machine learning: an exact penalty approach using a Difference of Convex function Algorithm , 2014, Machine Learning.

[32]  D. E. Gyi,et al.  Building healthy construction workers: Their views on health, wellbeing and better workplace design. , 2016, Applied ergonomics.

[33]  Takeo Kanade,et al.  Shape-from-silhouette of articulated objects and its use for human body kinematics estimation and motion capture , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[34]  Ayman Habib,et al.  Application of Microsoft Kinect Sensor for Tracking Construction Workers , 2012 .

[35]  Jun Li,et al.  Deeply Learned View-Invariant Features for Cross-View Action Recognition , 2017, IEEE Transactions on Image Processing.

[36]  M. S. Grewal,et al.  Application of Kalman filtering to the calibration and alignment of inertial navigation systems , 1991 .

[37]  Idsart Kingma,et al.  Optimal inertial sensor location for ambulatory measurement of trunk inclination. , 2009, Journal of biomechanics.

[38]  Mubarak Shah,et al.  Recognizing human actions using multiple features , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[39]  Kevin Huang,et al.  Development of a real-time three-dimensional spinal motion measurement system for clinical practice , 2006, Medical and Biological Engineering and Computing.

[40]  Abdulaziz M. Jarkas.  Effect of Buildability on Labor Productivity: A Practical Quantification Approach , 2016 .

[41]  SangUk Han,et al.  An Automated Biomechanical Simulation Approach to Ergonomic Job Analysis for Workplace Design , 2015 .

[42]  Jezekiel Ben-Arie.  The Probabilistic Peaking Effect of Viewed Angles and Distances with Application to 3-D Object Recognition , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[43]  Toby Sharp,et al.  Real-time human pose recognition in parts from single depth images , 2011, CVPR.

[44]  L Punnett,et al.  PATH: a work sampling-based approach to ergonomic job analysis for construction and other non-repetitive work. , 1996, Applied ergonomics.

[45]  Jochen Teizer,et al.  Real-time construction worker posture analysis for ergonomics training , 2012, Adv. Eng. Informatics.

[46]  Sakiko Oyama,et al.  Accuracy and repeatability of an inertial measurement unit system for field-based occupational studies , 2016, Ergonomics.

[47]  S Hignett,et al.  Rapid entire body assessment (REBA). , 2000, Applied ergonomics.

[48]  J H van Dieën,et al.  Stoop or squat: a review of biomechanical studies on lifting technique. , 1999, Clinical biomechanics.

[49]  Cordelia Schmid,et al.  Learning realistic human actions from movies , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[50]  Frédéric Bosché,et al.  Musculoskeletal disorders in construction: A review and a novel system for activity tracking with body area network. , 2016, Applied ergonomics.

[51]  Feniosky Peña-Mora,et al.  Empirical assessment of a RGB-D sensor on motion capture and action recognition for construction worker monitoring , 2013 .

[52]  SangHyun Lee,et al.  Computer vision techniques for construction safety and health monitoring , 2015, Adv. Eng. Informatics.

[53]  Pedro M. Domingos.  A few useful things to know about machine learning , 2012, Commun. ACM.