The main application of the technology is to increase safety and the protection of truck drivers during the container transfer process in a yard. This process is particularly prone to accidents as trucks, truck drivers, and straddle carriers are all operating in the same small area.
Before straddle carriers begin the pick-up/drop-off process with containers, drivers are expected to exit the truck cabin and wait until the carrier completes its task. The procedure requires an alarm to sound if the driver is not yet standing on a pressure-sensitive safety mat as the carrier approaches.
Camco has switched from its laser-based to image recognition-based technologies for this transfer process.
Previously, opening cabin doors blocking the view, operatives walking through an empty lane, or containers posed onto a surface would have prevented the laser-based solution to detect information needed to sound the alarm – or would cause too many false positives.
A vision-based solution with one Internal Protocol (IP) camera per lane for monitoring straddle carrier operations 24/7, the system is more responsive, recognising lane activity as soon as it is detected.
The vision-based solution is also robust against weather factors such as poor light conditions.
With deep learning of the CNN algorithms, the network will ‘learn’ by including these situations or images to its dataset.
Once the model is tweaked and tuned, it can be easily deployed and implemented for monitoring all the grids in the straddle carrier-operated container transfer zone.
Based on two CNN models, the central AI server is capable of processing and monitoring camera streams from up to 50 lanes in parallel.