Visual quality control of seeds, fruit and produce from images
Quality control comes down to a few questions asked of every sample: how many, how big, how uniform, what colour, and how many defects. These guides cover how a smartphone answers them from one calibrated photo — counting and sizing seeds and berries, grading apples by colour and ripeness, screening post-harvest seed lots — and where visual QC ends and lab or sorter-grade instruments take over.

Apple Colour Assessment with AI: Ripeness Checks, Starch-Iodine Tests, Leaf Area, Leaf Greenness and Disease Spots
How image analysis turns apple colour into a ripeness check, makes the starch-iodine test objective, and measures leaf area, leaf greenness and disease spot coverage in orchards.
21 October 2026
Grapevine Canopy AI Analytics: The New Core of the Viticulture Production Cycle
How AI-powered grapevine canopy analysis — Leaf Area Index, canopy cover, gap fraction, plus batch processing of images and video — turns smartphone photos into decisions across the whole viticulture production cycle.
14 July 2026
Counting Round Objects with a Smartphone: Blueberries, Seeds, Nuts and Computer Vision
How a smartphone counts round objects — blueberries, soybean and chickpea seeds, peanuts, beans, grapes — using computer vision and light deep-learning models. Get count, average size, and standard deviation from one photo, why 87% accuracy is often enough, and why a calibration plate is the rule for real measurements.
14 May 2026
Applied AI for Soybean Breeders: Tools, Workflows and Use Cases from the 2026 Soybean Breeders' Workshop
Dr Maryna Kuzmenko's talk at the 2026 Soybean Breeders' Workshop, written up in full: the difference between AI and applied AI for soybean, which imaging tool fits each growth stage from germination to post-harvest QC, why drones cannot count soybean flowers, and where a smartphone beats expensive equipment.
1 March 2026Petiole Pro runs visual QC on a phone: count, average size and standard deviation from one photo, with the annotated image exported alongside the numbers so every sample result can be traced back to what produced it.