Abstract
We consider an industrial internet of things environment, where involves multiple production factors, such as automatic guided vehicles (AGVs), people, container, etc. A deep learning model is presented for multi-target recognition, where the training data is shadow image formed by the nonuniform illumination of LED lighting source. Three shadow models of typical shapes are constructed to describe the shadows at different positions. The performance of the optimal VGG-16-based Faster-RCNN model is analyzed in view of the recognition accuracy and speed, and it is proved that recognizing three, four, and five types of objects, the mean average precision is 93%, 94.8%, and 92.6%, respectively. To enhance the generalization performance, the optimal Faster-RCNN is combined with the motion state of objects and the corresponding threshold. Simulation results show that the proposed deep learning model obtains significant performance gains to reduce missed and false detection.
Highlights
T HE industrial Internet of Things (IIoT) is an extension application of the traditional internet of things in the industrial field, involving all links and main parts of industrial manufacturing and internet communication technology
We propose a passive target recognition method, which uses the shadow formed by the blocked lights between the Lightemitting diodes (LEDs) and PD
We have proposed a passive target recognition method based on LED lighting
Summary
T HE industrial Internet of Things (IIoT) is an extension application of the traditional internet of things in the industrial field, involving all links and main parts of industrial manufacturing and internet communication technology. In the reflection-based technologies such as [13]–[15], the performance is affected by the signal-to-noise ratio (SNR) and the number of sensors These approaches only detect a single target at a time, and require plenty of LEDs and PDs, leading to more complex calculations. It is critical to develop a passive multiple targets recognition technology based on LED lighting. We propose a passive target recognition method, which uses the shadow formed by the blocked lights between the LED and PD. For the simulated light intensity distribution maps with shadows, the optimal Faster-RCNN model with high recognition accuracy and fast speed for multi-target is developed. Ar denotes the effective receiving area of the PD, Ts(ψ) is the gain of optical filter. gs(ψ) is the optical concentrator
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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
