Abstract
Telescopic cranes are powerful lifting facilities employed in construction, transportation, manufacturing and other industries. Since the ground workforce cannot be aware of their surrounding environment during the current crane operations in busy and complex sites, accidents and even fatalities are not avoidable. Hence, deploying an automatic and accurate top-view human detection solution would make significant improvements to the health and safety of the workforce on such industrial operational sites. The proposed method (CraneNet) is a new machine learning empowered solution to increase the visibility of a crane operator in complex industrial operational environments while addressing the challenges of human detection from top-view on a resource-constrained small-form PC to meet the space constraint in the operator’s cabin. CraneNet consists of 4 modified ResBlock-D modules to fulfill the real-time requirements. To increase the accuracy of small humans at high altitudes which is crucial for this use-case, a PAN (Path Aggregation Network) was designed and added to the architecture. This enhances the structure of CraneNet by adding a bottom-up path to spread the low-level information. Furthermore, three output layers were employed in CraneNet to further improve the accuracy of small objects. Spatial Pyramid Pooling (SPP) was integrated at the end of the backbone stage which increases the receptive field of the backbone, thereby increasing the accuracy. The CraneNet has achieved 92.59% of accuracy at 19 FPS on a portable device. The proposed machine learning model has been trained with the Standford Drone Dataset and Visdrone 2019 to further show the efficacy of the smart crane approach. Consequently, the proposed system is able to detect people in complex industrial operational areas from a distance up to 50 meters between the camera and the person. This system is also applicable to the detection of any other objects from an overhead camera.
Highlights
Telescopic cranes are widely employed in construction, oil and gas, maritime ports, transportation and manufacturing industries across the globe
The proposed object detection system was deployed and evaluated in a real-world scenario in an industrial site located in Stirling (Scotland), with access to a cameras attached to the hook of different cranes
The video feeds were sent from the customised Reolink camera (Reolink RLC-511W) attached to the crane hook (Fig. 5a) to the proposed convolutional neural network (CNN)-based model embedded on an NVIDIA Jetson Xavier small-form PC connected to a monitor for human detection through the Real-Time Messaging Protocol (RTMP) and Real-Time Streaming protocol (RTSP) (Fig. 5b)
Summary
Telescopic cranes are widely employed in construction, oil and gas, maritime ports, transportation and manufacturing industries across the globe. It is of myriad importance to improve safety by minimising the risks of crane operations. These crane-related hazards can be significantly reduced by Neural Computing and Applications industrial sites being cluttered and dynamically changing [10]. Our objective is to design and develop a new end-to-end real-time machine-learning-based human detection system (CraneNet) being able to detect people in cluttered industrial sites with high accuracy and efficiency considering the challenges involved in top-view human detection and more applicable to industrial use cases and especially crane sites. The application of the developed integrated system has been validated in real crane operation sites, such the one shown, located at Stirling, Scotland The application of the developed integrated system has been validated in real crane operation sites, such the one shown in Fig. 1, located at Stirling, Scotland
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
