right_division Green SCM Distribution
Bookmark us
SCDigest Logo

Focus: RFID and Automated Identification and Data Collection (AIDC)

Feature Article from Our RFID and AIDC Subject Area - See All

From SCDigest's OnTarget e-Magazine

- April 9, 2013 -


Supply Chain News: Could Visual Identification Technologies Revolutionize Logistics Operations?


Maybe DC Workers Could SimplyLook at an Item they are Picking Instead of Scanning It


SCDigest Editorial Staff


Tremendous progress is being made in the area of visual identification technology (VIT) -  with potential huge implications for supply chain and logistics systems.

The basic idea is this: increasingly, cameras and supporting software can be used to identify products in the same way a bar code or RFID tag might be used today. VIT systems identify products based on their image characteristic, and/or potentially specific markings.

SCDigest Says:


Such a system could even work in conjunction with regular bar codes. The worker could simply look at the bar code, a much faster process than needing to scan it.

What Do You Say?


Click Here to Send Us Your Comments
Click Here to See Reader Feedback

So called video or photo analytics are already starting to come to market in the supply chain. For example, at the National Retail Federation (NRF) trade show in New York in January, researchers from Carnegie Mellon University in Pittsburgh displayed a robotic system that helps manage "planogram compliance" in retail stores.

The camera-equipment robot can move around retail shelves and capture what the correct planogram (basically, how the store shelves are supposed to be set with different products) looks like, and then at whatever frequency the merchant wants the robot re-travels the aisles looking for variances from that plan that are then communicated to store managers or others for correction.

"Imager" scanners, which are essentially cameras, are increasingly making in-roads in supply chain and logistics applications too. In high speed, in-line applications such as a conveyor system, proponents of imaging technology say read rates can be much higher than traditional laser scanners, reducing problems from no reads. (See Is Imaging Technology Inevitably Going to Dominate Bar Code Scanning Applications?)

There is also really no reason, for example, that existing store cameras installed for security reasons could not also be connected to smart software that would send out an alert when a slot on a store shelf became empty and in need of replenishment. There are already systems that can tell if a specific individual, for example, interacted with a store associate and then whether that same person went through a POS lane for a purchase.

There have been great strides in facial recognition technology generally, picking specific individuals out in photos or video based on facial characteristics of those individuals stored in a database. That sure sounds a lot harder than distinguishing between various SKUs in a distribution center.

So consider something like this, for example: What if technology similar to "smart glasses" such as Google (see image below) was worn workers in a DC and connected to a mobile wireless terminal. Rather than perhaps scanning each object as it is selected and placed in a shipping carton or tote, is it possible that workers could simply look at the object and the system identifies and verifies what is being picked without any scanning at all? While smart glasses today are really about projecting web and other images to a person's eyes, we assume a sort of reverse model could be made, where images are captured and sent back to a device.


In fact, the web site of a company called Vuzix, a maker of smart glasses, notes on its web site an application where a person is "out shopping and spots an interesting item on the shelf. A quick snapshot of the barcode and your smart glasses enhanced app goes off to the Cloud, finds the product and competitive product and displays the data" - so that kind of image capture technology in glasses is here.

The beauty of that approach versus say RFID is that it would not require each product to have a still somewhat expensive tag on it.

Such a system could even work in conjunction with regular bar codes. The worker could simply look at the bar code, a much faster process than needing to scan it. That approach could be combined with a voice headset to allow the worker to correct any mistakes, such as "looking" at an extra bar code.

(RFID and AIDC Story Continued Below)




University of Arkansas Begins New VIT Study

In late March, two groups at the University of Arkansas announced a joint project to study ways in which visual identification technologies could be used in retail applications, as well as how VIT systems could be used to complement RFID technology.

According to the researchers there, VIT-based systems could be used to quickly recognize products on store shelves, add those goods to inventory lists, verify that they are at their correct locations and remove them from inventory upon checkout.


Source: Infographic by Martin Missfeldt

The university's Center for Advanced Spatial Technologies (CAST) has developed software that analyzes data gathered by two- and three-dimensional optical imaging hardware. The software, the university says, could be utilized for the purpose of geospatial location and mapping, as well as for shape analysis of objects, in order to help users equipped with cameras identify those items, as well as their locations in proximity with other nearby objects.

The research will focus in part on developing data standards for VIT systems in consumer goods and retail.
Justin Patton, the RFID Research Center's managing director, says currently there are no standard rules for storing visual data and matching that data to objects in a photograph. In other words, there is no standard set of descriptors or unique identifiers that would be utilized with each image of an object. If there were an open standard, retailers or other companies could share the standardized data related to an item.

Standards organization GS1 is the obvious candidate to develop such standards, but needs some of the spade work first in terms of this kind of research.

SCDigest believes it may not be too long before something like Google glasses may be coming to a DC near you.

Patton says there is a large and growing number of technology vendors working on VIT-based systems for the supply chain.

Do you see promise in VIT-based systems for the supply chain - or not? Let us know your thoughts at the Feedback section below.

Recent Feedback

ShelfSnap has been providing service in CPG and retail interpreting images into intelligence about assortment, out of stocks, facings and compliance. Our VIT-based system is quite different from those listed here in that it combines a variety of technologies and processes that recognize the chaos that is the retail shelf.

RFID and demand sensing technologies generate a large number of both false positives and false negatives.  We took the approach that we needed to be correct. 

Two final comments:

1. The math involved in interpretting many small products into a mosaic of information is quite different and more complex than facial recognition.

2. The biggest hurdle to the broadscale use of this in a consumer environment is the gap between the packaging that manufacturers and retailers think is on the shelf (and is therefore in the database feeding their application) and the package actually encountered by the consumer.  Our studies have indicated almost 60% of the products sitting on the Walmart and Kroger shelves are either not represented at all online or represented with the wrong image and data.

Mike Spindler
ShelfSnap LLC
Apr, 15 2013