Pallet scanning systems that read the label from any side

A pallet scanning system reads the shipping label on a pallet as it passes a fixed point, usually the dock door, so that nobody has to walk round the pallet with a handheld scanner. Built from cameras, it is a set of them around the door, one per face of the pallet. A detector finds the label wherever the wrapper left it, a reader decodes the barcode and the printed text from that crop, and every scan is logged with the frame it came from. The clerk's walk around the pallet disappears, and the labels the cameras cannot read are the ones the clerk is asked about.

Dock door 3 receives about forty pallets an hour on a Monday morning, and every one of them stops while the receiving clerk walks round it with a handheld scanner looking for the label. The shipper's dock crew put the label where it fitted, the wrapper wound two turns of film over it, and the forklift driver set the pallet down with that face against the wall. The clerk finds it eventually. Forty times an hour, eventually adds up to a forklift idling at the door for most of the shift.

The dock cameras were installed for the insurance claims. They already see every face of every pallet that comes through the door.

How a pallet scanning system works

There are four stages, and the first three run in the second or two the pallet takes to pass the door.

The first is capture. Cameras on each side of door 3 take frames as the forklift drives through, so that every face of the pallet is seen by at least one of them while it is in the light and in focus.

The second is finding. A detector puts a box on every shipping label in every frame, whichever face it is on and whatever angle the film has pulled it to.

The third is reading. The barcode inside each box is decoded, and the printed text beside it, the shipper, the purchase order, the item count, is read into fields.

The fourth is the handover. The decoded label is matched against the receipts the warehouse system expects today, the match or the mismatch goes to the dock office, and the frames are kept with it.

The dock already owns the first stage. The receiving clerk is the second, third and fourth today, with a handheld scanner and a clipboard.

The label lands wherever the wrapper left it

There is no standard face at door 3. A pallet built at one supplier carries its label on the short side at chest height, and a pallet from another has it low on the long side under the film. A mixed pallet from a cross-dock carries three labels from three legs of its journey, and only one of them is the one this dock needs. A rule that says "the label is here" is wrong on the next pallet through the door.

The package and label inspection use case describes the same problem on a packing line, where every check is a region and a rule: find the label, then decide about it. At the dock the finding is the hard part, because the region can be anywhere on four faces.

A portal of cameras sees every face at once

The portal is the dock's existing cameras plus two more, one each side of the door at label height, so that a pallet passing through is seen on all four faces in the same few seconds. On every frame a model boxes the labels, all of them, whichever face they are on and whatever angle the film has pulled them to. The class is typed once, label, and Lexi proposes the boxes on frames the dock has already recorded. The receiving clerk checks them, and his attention goes to the labels seen through film, at a slant, half in shadow, because those are the ones the handheld scanner could not read either.

A second class, wrapped label, is boxed separately for the labels the film has crossed, because what happens next is different for those.

The barcode is decoded and the printed text becomes fields

Inside each box is a barcode and a block of printed text. On a pallet that follows the GS1 logistics label, the barcode is a GS1-128 carrying the Serial Shipping Container Code, the number that identifies this one pallet through the chain. It is eighteen digits long, it always sits behind the application identifier that marks an SSCC, and the same digits are printed in human-readable form under the bars (GS1 UK, logistics labels and SSCCs).

That second copy is what makes the camera version of the portal different from a barcode reader. When the film has crossed the bars and the decode fails, the printed digits beside them are read instead, and the number is matched against the advance ship notices for the day.

LexInsight takes the reading and turns it into fields the warehouse system can match, and the question "which pallets arrived against purchase orders we were not expecting" has an answer with the frames behind it.

Reading text through two turns of stretch film is not reliable, and the system should say so rather than guess. A label the reader cannot make out goes to the clerk with the frame, and the clerk types the number from the screen, which is faster than walking round the pallet and is still the exception rather than the rule.

Every scan is logged with the frame the portal saw

Each pallet through door 3 leaves a record: the label boxes and the decoded barcode, the fields that were read, the time, and the frames from all four cameras. When the shipper disputes a short delivery in three weeks, the record has the frames of the pallet as it came through the door, with the label the clerk would have been asked to find.

On the docks we work with, that record replaces the photograph the clerk used to take with a phone when something looked wrong, and it exists for every pallet rather than the ones that looked wrong at the time.

Cameras, barcode gantries, handhelds and RFID compared

A camera portal is one of four ways to take the scan off the clerk, and the right one depends on what the dock needs to keep. The table is the comparison as it stands at a receiving door.

Method	What it reads	What every pallet needs	Film over the code	The forklift	What is kept
Handheld scanner	The barcode, one at a time	A label within reach	Read only where the film is clear	Stops while the clerk finds the label	The decoded number
Fixed barcode gantry	Barcodes on the faces its readers cover	A label on a face a reader sees	Clear film is readable, a fold across the bars is not	Drives through without stopping	The decoded number and a trigger event
RFID gate	A tag, not a label	A tag on every pallet, tested for the load	Film is no obstacle	Drives through without stopping	The tag's number, with no picture of the pallet
Camera portal	The barcode, the printed text and the pallet itself	The label it already carries	The printed digits beside the bars, or the clerk with the crop	Drives through without stopping	The decoded fields and the frames from every face

The gantry is the established answer where the labels are consistent and the volume is high. Zebra's dock door scanning mounts fixed readers on a gantry around the door so that forklifts pass through without a stop, reading codes at several distances and angles on the faces the readers cover (Zebra, dock door scanning). What it keeps is the number. What it cannot do is read the human-readable digits when the bars are creased, or show the dock office what the pallet looked like.

RFID moves the problem from the label to the tag. A tag on every pallet means a tag applied at the shipper or at the door, and a load of metal or liquid changes how the tag reads. One logistics provider found that mixed pallets of metal, liquid and foil items would not meet its retailer's read-rate requirement on standard passive tags, and changed to a different tag type for those loads (RFID Journal). An RFID gate at door 3 would read through the film, and would know nothing about the three-label cross-dock pallet or the short delivery dispute, because it never saw the pallet.

The camera portal is the answer where the labels are inconsistent and the dock wants the picture. The dock door utilization post uses the same yard cameras for a different question, how long each trailer sits at each door, and the pallet accounting post uses an overhead camera to count pallets through a zone. All three run on the cameras the building already has.

Building the portal in six steps

The build is the same shape as every other camera on the site, and most of it is deciding what the labels mean before anything trains.

1. Look at the door through the cameras it has. Pull a frame from each dock camera with a pallet in the door and find the label in it. If a person cannot read the human-readable digits under the barcode on the frame, the reader will not either, and that face needs a camera closer to label height. Door 3 needed two, one each side of the frame.
2. Record a week of pallets and pull the frames. The recorder already has the footage. Converting DAV footage to MP4 is the usual first job when the recorder is a Dahua, and the frames come out of the MP4 at the sampling rate the portal will run at.
3. Label two classes. Type "label" and Lexi puts a box on every shipping label in every frame; the clerk reviews them, with the slanted and shadowed ones getting the time. A second class, "wrapped label", goes on the labels the film has crossed. The labeling guide describes the review pass, and the rule for a label half hidden by the forklift mast is written down before anyone labels one.
4. Train and check on a day the model never saw. The check is on pallets, not boxes: for each pallet through the door, did at least one face produce a decoded number that matched a receipt. The pallets that did not are the list the portal would have sent to the clerk.
5. Write the rule. A decoded number that matches an expected receipt is logged. A number that matches nothing goes to the dock office with the frames. A label the reader could not make out goes to the clerk with the crop and the expected receipts. The monitoring and alerts guide covers where each of those goes and how often.
6. Run it beside the recorder. The runner sits on the warehouse network, the frames stay in the building, and only the doubted crops leave it. The deployment guide describes what leaves and what does not.

The second version of the model comes from step five, because the crops the clerk typed from in the first month are the labels the first version was unsure of.

Stretch wrap over the label needs a physical fix

The failure to plan for at door 3 is film. A wrapper that winds one more turn than usual puts a fold of film across the barcode, and the frame is fine, the label is found, and the code will not decode. The drift catalog calls this something is now in the way, and its advice is the right one at the dock: check the frame before you retrain. A model retrained on frames of wrapped labels learns to expect the fold, and no amount of training reads a barcode through it.

The fix is at the shipper. The wrapped label class exists so the dock can count how many pallets from each supplier arrive with the label under film, and that count, per supplier per week, is a conversation with the supplier's dock crew rather than a model problem. The same count, taken at the dock's own packing line, is the subject of label inspection before the case is sealed.

The portal decides on site, beside the dock's recorder

My view, which comes from where things run rather than from the model, is that a dock portal should run on a runner on the warehouse network, beside the recorder the cameras already feed, rather than in the cloud alone. A pallet at the door waits for the decode, and a decode that has to leave the building and come back is a forklift idling on the network's schedule.

The portal is one camera of several in the building, and computer vision in the warehouse, dock door to sorter walks the rest of them. On the runner the label is found and decoded on site, the footage stays where it was recorded, and only the frames the model doubts leave the building.

LexData takes the portal's model through its whole life. You type what to look for, Lexi puts a box on every label in every frame, and a person checks each label before anything trains on it. The model then watches the dock cameras, in the cloud, on your servers or on a runner beside the recorder. Frames it is unsure of come back to a person, the corrections retrain it, and the new version replaces the old one with no downtime. The slanted, shadowed labels the clerk typed by hand in the first month are the second version's evidence, and the forklift at door 3 stops idling.

Four questions dock teams ask about pallet scanning

Does a camera portal replace the handheld scanner

No. The handheld stays in the dock office for the exceptions: the pallet with three labels where the rule picked wrong, the label the portal sent to the clerk, the pallet that came in through the personnel door. The portal takes the forty pallets an hour, and the handheld takes the two the portal asked about.

Can the cameras read the barcode through stretch film

Through clear, flat film, usually. Through a fold or a crease across the bars, no, and nothing reads a barcode through a crease. What the camera portal has that a barcode reader does not is the printed digits beside the bars, which survive a crease more often than the bars do, and the clerk with the crop when they do not.

Which label counts on a pallet that carries three

The one whose number matches a receipt the dock expects today. The other two are logged with the frame, because the dispute three weeks from now is sometimes about the leg of the journey the dock did not know it was on.

Does the dock need new cameras

Usually two, one each side of the door at label height, because the security cameras were mounted high to see faces and a label at knee height on the far side is a smear from there. Everything else is the cameras and the recorder the building already has.