BRENNEN

Industries / Food, farming and forestry / Food and beverage processing

Seafood processing

Seafood processing turns landed fish and shellfish into fillets, frozen blocks, cans, smoked fish and other products, and puts what is left over to use.

About 9,200 years ago

At Norje Sunnansund in southern Sweden, people stored freshwater fish by fermenting it in a pit, without salt. Adam Boethius of Lund University titled his 2016 paper on the find "the world's earliest evidence of fermentation".

eurekalert.org

39

Milestones achieved across 6 domains.

12

To watch for: announced, not yet confirmed outside.

12

Open deliverables. Nobody has shipped them.

In one minute

Where it started

People have dried, salted, smoked and fermented fish for thousands of years, and a pit in Sweden held fermented fish about 9,200 years ago. Factory processing began in the 1800s, with sardines canned in oil at Nantes from 1824 and salmon canned on the Pacific coast from 1864. In the early 1900s machines took over gutting salmon and filleting herring, and Clarence Birdseye's patents of 1924 to 1930 froze fish in pressed blocks.

Where it stands

In 2024, by FAO's count, 34 percent of the fish and shellfish people ate was sold frozen, 12 percent prepared and preserved, such as canned, and 10 percent dried, salted, fermented or smoked; the rest was sold live, fresh or chilled. Machines fillet whitefish and salmon, peel shrimp, cut out pin bones and freeze at scale, but finding worms, trimming fillets and opening live oysters are still done by hand, and much fish is thawed, cut and refrozen in another country. Hazard plans, catch certificates and DNA tests now follow the fish, but reporters found forced labour in peeling sheds in 2015 and in Chinese plants in 2023 that the paperwork did not show.

What is open

Most open jobs are checks that people still make by eye, or by destroying the sample: grading freshness, finding worms, telling twice-frozen fish, weighing ice glaze, measuring can seams and screening smoked fish for smoke chemicals. Low-cost cameras, handheld spectrometers and image models that train on one graphics chip make these worth attempting now, as public test sets and bench prototypes. Others are missing public data: labels checked against DNA, what each fish leaves behind, and where one port's leftovers go.

Domains

AchievedWatch forOpen

Figureheads

1 person

Open in seafood processing

3 of 12 open deliverables

Open · Sorting and grading

An open worm image set

A public set of cod or hake fillet images, taken on a light table with an ordinary camera and a low-cost near-infrared camera, with every worm's position confirmed by the standard laboratory method, and a worm detector tested on it.

Built with what exists

A light table, a colour camera, a low-cost near-infrared camera such as the Raspberry Pi Camera Module 3 NoIR with an infrared lamp, the UV-press method of ISO 23036-1:2021 (fillets pressed thin and frozen, then viewed under ultraviolet light, where the worms glow) run with a food laboratory for the reference count, and an object detector such as Ultralytics YOLO11, which trains on one graphics chip.

Credit · unclaimed

Open · Filleting, peeling and shucking

Mapping the oyster's muscle

A public set of oyster 3D scans or photos with the hinge and the adductor muscle, the muscle that holds the shell shut, marked after hand shucking, and a model that predicts from the outside of the shell where that muscle sits. A machine that opens live oysters needs this map first.

Built with what exists

A phone camera or a structured-light 3D scanner such as a Revopoint POP 3, an oyster knife and a cut-resistant glove for the reference shucking, Meta's open Segment Anything 2 model to outline shells, and a small model trained in PyTorch.

Credit · unclaimed

Open · Freezing

Spotting twice-frozen fillets

A quick check that sorts whitefish or salmon fillets into fresh, frozen once and frozen twice without cutting them, using a handheld near-infrared spectrometer, tested on fillets whose history is known.

Built with what exists

A handheld near-infrared spectrometer such as Si-Ware's NeoSpectra Scanner or VIAVI's MicroNIR OnSite-W, fillets with documented freezing histories from a processor or a home freezer, and Python with scikit-learn for the classifier.

Credit · unclaimed

Where it sits