Author

Finn Ellingwood

Engineer · LexData · 27 articles

Packaging line, cartons with labels passing a scanner, carton and label boxed, generated scene with detections from our model

Computer vision · 6 min read

What a convolutional neural network does with a frame from a line camera

One small filter slides across every position in the frame, and that single design decision is why the network fits on a box beside the recorder.

Finn Ellingwood · Oct 4, 2026

Checkout lane from a ceiling camera, a person and the products on the belt boxed, no faces, generated scene with detections from our model

Computer vision · 7 min read

Object counting with computer vision, three kinds of count and why each needs its own pipeline

Forty boxes on a pallet, unique trucks through a gate over a shift, people in a checkout zone right now. Same word, three pipelines, and two ways to be wrong.

Finn Ellingwood · Oct 3, 2026

Cartons with labels moving past a scanner on a packaging line, carton and label boxed, generated scene with detections from our model

Computer vision · 7 min read

OCR explained, from a lot code label to a verified read

Optical character recognition on a packaging line camera, which is not a scanner. Acquisition, detection, recognition and validation on a lot code label.

Finn Ellingwood · Oct 3, 2026

Assembly station with a housing, fasteners and a cable laid out, the cable boxed, generated scene with detections from our model

Labeling · 7 min read

Polygon object detection when the model only draws boxes

Polygon object detection, or polygons as labels for a model that draws boxes. A diagonal part gets a box that is mostly bench; a polygon keeps it tight.

Finn Ellingwood · Oct 3, 2026

Fillets on a blue processing conveyor, each one boxed where they lie against each other, generated scene with detections from our model

Labeling · 6 min read

Instance segmentation data labeling where two parts touch

Two fillets overlap on a blue conveyor and a corroded patch fades into clean steel. The label is a decision about where one thing ends, made before the drawing.

Finn Ellingwood · Oct 2, 2026

Edge box beside the recorder in a plant cabinet, generated scene with detections from our model

Edge · 7 min read

What the ONNX file is, and why it sometimes loads and returns nothing

One exported file runs on the Jetson at the line, the Pi at the second site and the rack server. When it loads and draws no boxes, check the operator set first.

Finn Ellingwood · Oct 2, 2026

Packaging line, cartons past the scanner, generated scene with detections from our model

Labeling · 7 min read

Uploading a week of images, videos and annotations without losing the rare frames

A week of the scanner camera is mostly the same carton. Import the originals, sample the routine, keep the jam at 3 am whole, bring old labels across intact.

Finn Ellingwood · Oct 2, 2026

A checkout lane from a ceiling camera, person and products boxed, generated scene with detections from our model

Operations · 7 min read

Queue counting with a polygon on the checkout camera

Queue counting from the camera over the checkout. Draw the polygon once, keep the right class, test the bottom of each box, and alert on a held count.

Finn Ellingwood · Oct 1, 2026

Packing bench with boxes and labels boxed, generated scene with detections from our model

Operations · 7 min read

QR code detection with computer vision when the code on the bin will not read

Finder patterns and the quiet zone for whoever is debugging a failed read, a detector that crops the code from a wide frame, and the result logged per tote.

Finn Ellingwood · Oct 1, 2026

Cartons passing a scanner on a packaging line, generated scene with detections from our model

Operations · 6 min read

Machine vision lens selection worked back from the smallest defect in pixels

The lens is chosen from the defect, not from the drawer. Start with how many pixels a torn seal needs, and a later lens swap is a new camera to the model.

Finn Ellingwood · Sep 30, 2026

Parked vehicles boxed beside driveways from above, landing areas marked, from a customer drone run

Industries · 7 min read

Parking lot occupancy detection from the camera already over the car park

A region per bay, occupied or empty as the call, a free-bay count for the sign and the manager, and snow, glare and a nudged lamp post as what goes wrong.

Finn Ellingwood · Sep 30, 2026

Orchard edge from a customer tractor camera, tree rows and the drivable path boxed

Industries · 7 min read

Plant height measurement with a camera and a stake of known height

Plant and stake as two classes, pixels to centimetres through the stake, a growth curve per plot from a question, and wind and a nudged mount absorbed by it.

Finn Ellingwood · Sep 30, 2026

Distribution poles along a desert track, one insulator flagged as cracked, from a customer inspection run

Operations · 6 min read

Slicing the frame so the small things get found

A cracked insulator is a few dozen pixels in a frame thousands wide, and the detector shrinks the frame before it looks. Tiles keep the pixels, at a price.

Finn Ellingwood · Sep 29, 2026

Drone pass along a solar farm, panels boxed across the rows, from a customer drone survey

Industries · 6 min read

Solar panel detection in aerial imagery without counting the skylights

Over a suburb a panel is a few dozen pixels in a dense grid and a skylight is the same dark rectangle. Box per array, tile the frame, and keep the snow days.

Finn Ellingwood · Sep 29, 2026

Bread shelf with an empty slot flagged and the rack sections boxed, from a customer store camera

Labeling · 6 min read

AI labeled vs human labeled data, how much a model can do before a person has to look

On a shelf dataset the model's proposals are mostly right. On safety cones in a yard they are mostly wrong. Measure the gap, then keep a person on every label.

Finn Ellingwood · Sep 28, 2026

Street scene from a vehicle camera, eleven traffic signs, pedestrians and vehicles boxed, from a customer perception run

Computer vision · 7 min read

What is DETR, the detection transformer explained

DETR, explained. A fixed set of learned queries instead of anchors, no suppression step, and the small-object and slow-training problems later work fixed.

Finn Ellingwood · Sep 28, 2026

Compute box beside the recorder in a plant cabinet, recorder and cables boxed, generated scene with detections from our model

Edge · 7 min read

Running computer vision on RTSP camera streams from the recorder you already have

A dozen IP cameras and one NVR. One consumer per stream, a frame about every two seconds, TCP when the switch is busy, and a bitrate change ruled out first.

Finn Ellingwood · Sep 27, 2026

Bottling line with bottles queued under the fill head, each boxed, generated scene with detections from our model

Operations · 6 min read

Counting objects on a conveyor belt without counting any of them twice

Bolts tracked across frames so each is counted once, touching parts as the label to settle, and a compression change as the step that breaks the count.

Finn Ellingwood · Sep 26, 2026

Stamped steel panels on the conveyor after the press, generated scene with detections from our model

Industries · 7 min read

Dimensional defect inspection at the press outfeed, for the bracket that looks perfect and will not fit

Hole spacing, gap width and alignment measured on every part against tolerance, a drifting tool offset caught as a slope, and a new die labeled before the run.

Finn Ellingwood · Sep 26, 2026

A site yard with a scaffold tower beside containers, workers and ladders boxed, from a customer site camera

Industries · 7 min read

Rebar counting from one photo of the bundle end, with a bounding box on every bar and a record for the ticket

A camera over the saw end of the bundle, a box on every bar, the count checked against the delivery ticket, and a re-mounted camera as the drift to expect.

Finn Ellingwood · Sep 26, 2026

Robotic weld cell with a metal part on the fixture, generated scene with detections from our model

Industries · 7 min read

Body-in-white inspection with a station camera on unpainted steel

Dents on a bare body shell hide in the reflections. Masks give the extent, a second pass decides reportable, and a tightened tolerance is a spec change.

Finn Ellingwood · Sep 25, 2026

Steel sheet passing an inspection station on a stamping line, generated scene with detections from our model

Industries · 8 min read

Semiconductor wafer inspection with a bright-field camera at intake

A cracked wafer rejected at intake costs a wafer. Found after lithography it costs every step spent on it. Three classes, and masks where extent decides.

Finn Ellingwood · Sep 25, 2026

Assembly bench with a tool, a tray and wiring boxed, generated scene with detections from our model

Industries · 7 min read

Manual assembly QA with a camera that catches the wrong part in the assembler's hand

A camera over the bench boxes and names the part in hand, checks it against the next step, and lights a lamp before the next screw goes in.

Finn Ellingwood · Sep 25, 2026

Transmission lattice tower from a drone pass, corrosion flagged on the hardware, from a customer inspection run

Industries · 6 min read

Telecom tower inspection for corrosion, measured as area from a drone pass up the legs

Rust on a lattice tower as a mask so two flights can be compared, the member it sits on as the priority, and salt haze as the frames the model has seen least.

Finn Ellingwood · Sep 24, 2026

Robot arm loading parts into a CNC machine behind a safety fence, generated scene with detections from our model

Industries · 6 min read

Vision guided robotics, giving the robot a reason to look before it moves

A casting a few millimetres off its nest stops a blind program. The camera-guided one finds the part, estimates its pose and adjusts, until the mount is bumped.

Finn Ellingwood · Sep 24, 2026

Two workers under scaffold netting with hard hats boxed, from a customer site camera

Industries · 6 min read

PPE detection in production, deciding who is wearing the hard hat

A gate camera, three classes and a spatial rule decide compliance per person. The hard hat carried in a hand is the label that decides whether the model holds.

Finn Ellingwood · Sep 23, 2026

Robot arm loading parts into a CNC machine behind its safety fence, fence and arm boxed, generated scene with detections from our model

Industries · 6 min read

Vision AI for factory robots, picking from a bin nobody arranged

A cell camera over a bin of oily sheet metal blanks, each found with its position and rotation for the arm, and the new part number labeled before changeover.

Finn Ellingwood · Sep 23, 2026