Machine vision: IoT at the edge in action
ADLink demos how machine vision and edge compute drive efficiency and savings for palletization
Virtually any company that makes, sells or transports goods spends time and money putting boxes onto pallets. While this function is easy to overlook, it’s currently a largely manual process that takes time, money while adding risk. However, as companies increasingly digitize operations and take advantage of data analysis and insights, palletization is set to take a step forward with machine vision.
So what is machine vision. Think of the larger internet of things–networked sensors generating massive volumes of data that, if utilized correctly, can create some sort of process or monetary efficiency. Now narrow in on sensors–whether it’s air quality metrics, traffic volume over time, number of times a facility was accessed or imagery. In that latter case, a camera effectively serves as a sensor. And if you layer in software tools, that imagery can lend the machine (a camera and the system it’s attached to) vision.
Let’s take a look at an example of how the combination of machine vision and edge compute power augment human abilities by accelerating the palletization process and reducing errors. Instead of hand-scanning a code on a box, a handler holds the box underneath an array of cameras that read the information and determine whether the box is about to be placed on the correct pallet. If the answer is yes, the machine vision system gives the human worker a visual cue, in this case a flashing green light. If it’s not the correct pallet, the worker sees a flash of red light.
This puts a fine point on the importance of edge computing. You could do the same thing using centralized compute power but it would take a lot longer for the machine vision system to make a determination. In latency-sensitive use cases like this, compute has to be localized.
ADLink demoed this combination of technology in Las Vegas during the recent Pack Expo.
“Our Edge IoT solutions can help make conveyors smart, find missing inventory, automate bin picking, determine fill levels, connect robots and more − all in real-time,” says Daniel Collins, ADLINK IoT Director for North America. “By bringing AI to the edge we’re helping to automate warehouse logistics in a quick and cost-effective way that increases productivity and employee ergonomics. One of our customers decreased the time it takes to build a pallet by 41%, increasing total daily throughput by 200% without disrupting the way employees are used to working.”