Seed colours (testa and cotyledon), patterns, size and shape, and the within-variety consistency of these characteristics, are key determining factors for quality and price of lentil. These are also key traits for variety identification and are required for variety registration and for establishing distinctness, uniformity and stability (DUS) when applying for Plant Breeders’ Rights. In most lentil breeding programs, visual quality traits are evaluated by laying out thousands of samples for visual evaluation. Seed size and shape are determined via a time-consuming procedure using a set of graduated round-hole and slotted sieves. An automated, objective measurement of these seed characteristics is needed to improve efficiency and throughput of the breeding program.

Computer image processing can alleviate the weaknesses of traditional phenotyping by measuring seed traits very accurately, including both morphological and color characteristics. This high-throughput technique has been proven to be very powerful and reliable to measure traits such as seed dimensions in lentil. Under the Plant Phenotyping and Imaging Research Centre (P2IRC; a $37.2M project funded by the Canada First Research Excellence Fund), a high- throughput imaging device - BELT (Better Evaluator of Lentil Technology), was built by engineers in S. Noble’s group to image lentil seeds. It currently uses image capture and analysis to measure shape, size, and colour properties on individual seeds and return descriptive statistics. Within-sample variation will give an indication of the uniformity of samples, while data collected for a population will allow us to characterize phenotypes for later use in QTL or GWAS analyses.