Computer vision technology is rapidly replacing the human eye in logistics sorting processes. Cameras and software powered by artificial intelligence (AI) are already reading package labels, addresses, handwritten text, and barcodes with incredible accuracy and speed. The widespread availability of high-quality cameras and graphics processing units (GPUs) gives operators a deeper understanding of warehouse processes, not just what’s on the conveyor belt. The benefit? More efficiency, fewer errors, and easier work.
Lorenzo D’Arsie, Computer Vision Product Manager and Lars Pruijn, Innovation Director at Prime Vision, examine computer vision technology and the new opportunities it provides in the postal and parcel sectors.
Vision capabilities move beyond OCR
Currently, computer vision technology in warehouse logistics is primarily focused on optical character recognition (OCR) to improve sorting efficiency, accuracy, and throughput. OCR captures still images from a camera mounted above a conveyor belt and can read and understand labels, addresses, barcodes, symbols, and even handwritten or printed text. This information is used by humans, sorting machines, and robots to get goods to the right destination. Products such as Prime Vision’s Address Vision and Barcode Vision are powered by AI to reconstruct and interpret damaged or unrecognizable packaging information.
Computer vision is great, but it’s bringing so much more potential these days. Cameras with higher resolution, improved color fidelity and improved 3D depth of field allow for higher quality images and videos. GPUs provide the computing power needed to process this growing amount of data, while deep learning techniques for text and object recognition enable more efficient quantification. What’s interesting for warehouse managers is that these technologies are becoming more commercially available and less complex to implement.
Setting the scene for greater understanding
These capabilities allow warehouse workers to see beyond the conveyor belt. What if warehouse employees could see and link every event of a package moving through the facility, rather than just getting one chance to read a letter or package label?
This is called “scene understanding” and it can reduce errors even in the most controlled sorting environments. Combining all the information at the network level allows issues to be identified earlier and the entire process to be evaluated. This allows for proactive thinking and better handling of exceptions that can drive up costs for the company.
So how does this work in practice? Computer vision software has access to every camera installed in your facility and has an innate understanding of each unit’s relative location. This intelligence reduces calibration efforts and simplifies setup. This expanded field of view allows the system to identify problems and help resolve them.
For example, the system can help detect packets that could not be read. It can also detect overlapping or stuck-together packages. It can also highlight objects that cannot be machined by assessing their size, shape and instability, quickly sorting them before they cause issues or damage equipment. For businesses with a wide variety of goods, such as the postal market, this streamlines the handling of different packages and saves time.
Beyond the parcel stream, computer vision can track roller cages to confirm they reach the right destination. The system can see whether doors are open or closed, pointing out where improvements to efficiency or safety could be made. Cameras on loading bays can monitor trucks moving in and out, allowing delivery trends to be analysed and further logistical insights gained.
Furthermore, computer vision can identify when ergonomic rules aren’t being adhered to, an important safety factor in areas with machinery and heavy items. New efficiency opportunities and best practice can be identified and applied in day-to-day work, making certain tasks easier for the workforce. Across the warehouse, computer vision is a powerful tool for improvement.
Computer vision in real warehouse applications
As an expert in computer vision, Prime Vision is harnessing this technology in real-world applications, designing and implementing customised systems for specific tasks.
At one Prime Vision customer, operators were using a large parcel sorting machine inside a casing. Over time, debris collected within the casing, which would require the machine to be stopped so that maintenance personnel could conduct a visual inspection to find and remove it. This was a time-consuming process. To alleviate this, Prime Vision installed a computer vision system underneath the machine that could quickly detect and locate any debris or small parcels that had fallen into the casing. Consequently, removal could be completed faster by staff, helping to promote uptime.
An internal research project involved using computer vision to support a manual sorting system. A complex sorting process was simulated, involving non-conveyable items being moved manually to 40 different locations – an approach that often results in a high number of sorting errors. Prime Vision tested a system to check if specific items were placed in the right cage and provide alerts if an item ended up in the wrong place. Results showed the system’s potential to dramatically reduce errors.
Computer vision systems can enhance the operation of a fully automated system too. Many warehouses are trialling the use of robot arms to load individual parcels onto conveyor belts in a logistical equivalent of pick and place. However, as a relatively new application in the sector, sometimes robots can pick up two items at once by accident, causing sorting issues. Computer vision can fix this by identifying when this occurs, providing alerts to operators or the robot, optimising the process.
Eyes on the future
Clearly, advancements in computer vision and the increasing ease of access will enable warehouse operators to better monitor, understand and optimise warehouse sorting processes. There are some considerations going forward though.
First is that warehouses are inherently conservative environments, where adoption of new technologies will be gradual. For example, many operations will continue to run off central processing units (CPUs) instead of GPUs because of the high upfront hardware costs of changeover. Big infrastructural changes won’t happen overnight, but the capability is definitely here today.
The other factor is privacy. Video surveillance and access to personal images is a complex subject, so any computer vision system needs to be focused solely on tracking objects and processes, not people. There are various solutions to achieve this, such as blurring out images, using black box AI systems with no visibility or positioning cameras accordingly.
If these conditions are met, computer vision offers the possibility to act as a helpful assistant to warehouse operators, making work easier and crucially, more efficient.