satinder-pbg@pau.edu

Wheat (Triticum aestivum) stands as a cornerstone of global food security, yet confronts escalating pressures from climate change and population growth, necessitating the expedited development of high-yielding, climate-resilient, and nutritionally enhanced wheat varieties. The upgradation of wheat breeding system for catering to these threats may be structured around two functional units- one aimed at acceleration of the genetic and breeding processes and the other at high throughput precision phenotyping using artificial intelligence tools.

To address the first component, a cost-effective speed breeding chamber, measuring 3 m × 4.5 m × 3 m, was constructed using Polyurethane Foam (PUF) panels, strengthened by eight 100W and four 200W LED light boxes and two 1.5 hp split type domestic air conditioners. This innovative chamber operates on a 7KW solar panel system, highlighting a novel approach to accelerate plant growth and development via solar-powered artificial lighting and temperature control.

Using these advancements, a ‘Recombinant Inbred Line’ (RIL) population, derived from a cross between 'PBW 1 Chapati' and ‘Aegilops kotschyii introgression line’, was rapidly developed under optimized growth conditions, culminating in a 365-to-377-day developmental cycle of five generations. The F6 generation from this was under field evaluation for precision phenotyping in the main crop season 2023-24.

The duration of the seed-to-seed cycle ranges from 72 to 83 days, depending upon the genotype involved for accelerated breeding in the low-cost speed breeding chamber. Alongside, hybridization has also been prioritized in the speed breeding chamber and fully viable crossed seeds during the summer, when outdoor temperatures range between 38°C to 45°C were harvested, showcasing the chamber's adaptability to diverse breeding exigencies. This accelerated development of germplasm necessitates the integration of novel precision phenotyping tools to expedite the delivery of elite varieties to farmers.

The cost-effective, user-friendly algorithm and workflow tailored to stage- and stress-specific phenotyping consisting ofa computer algorithm was devised to determine tiller numbers, grain dimensions, and morphology, using size and shape descriptors including minimum bounding rectangle (MBR), area (A), perimeter (P), solidity, minor diameter (m), and major diameter (M).

Employing digital image processing techniques, this algorithm, developed using the open-source Python language and OpenCV library, promises to revolutionize high-throughput field phenotyping, facilitating the characterization of advanced germplasm through big-data methodologies. Combined incorporation of speed breeding with precision phenotyping tools holds a transformative shift in wheat breeding towards sustainable agriculture and food security.