Defects in french fries come in many forms: black spots, green ends, bruises, discoloration, skin defects and rotten parts. S-Blade detects each one with deep learning and removes only the defect and preserves the rest of the french fry.
Black spots
Black spots appear as small, dark patches inside or on the surface of the fry. S-Blade detects them by colour and shape and cuts out only the affected section. This preserves yield while removing a visual defect that would otherwise lead to a full reject.
Green ends
Green ends form when potatoes are exposed to light and develop chlorophyll. Thanks to deep learning, S-Blade removes even the smallest, light green ends, cutting them out while the rest of the fry continues down the line.
Rotten parts
Rotten and soft-rot areas look darker than healthy potato. Removing them matters for food safety and end product quality. S-Blade's deep learning model distinguishes rotten tissue from acceptable variation and removes the defect. The healthy part of the fry stays on the line, so removing a serious defect does not mean losing the whole piece.
Discoloration
Internal and external discoloration can come from heat damage, bruising or storage conditions. The affected zone is often small but visually significant enough to downgrade the fry. S-Blade identifies the discoloured section by colour and removes only that part, preserving the surrounding good product and keeping the end product visually consistent.
Skin defects
Skin imperfections and scab marks are common in skin-on fries. Thanks to deep learning, S-Blade can see the difference between healthy skin and a skin defect: it removes the defect while keeping the healthy skin intact.
Bruises
Bruises are internal damage that often only becomes visible after frying, when the colour changes during cooking. That makes them hard to catch with simple surface inspection. S-Blade detects bruise patterns using deep learning, and removes them before they reach the fryer, preventing discoloured product in the finished batch.
