Seeds

Agricultural research has accelerated in recent years, but farmers often lack the time and resources to conduct on-farm experiments, as most of their efforts are devoted to crop production. Seed classification provides essential insights for seed quality control, impurity detection, and yield estimation. Early identification of seed types is critical to reduce costs, minimize risks of poor field emergence, and support efficient crop management. Traditional classification methods rely heavily on manual feature extraction and expert input, which limits scalability and accuracy when dealing with highly similar seed types. To address this challenge, we propose an automated end-to-end deep learning framework for complex multiclass Brassica seed classification. Our framework integrates preprocessing, feature learning, and classification into a unified pipeline, eliminating the need for handcrafted features. Using a newly collected dataset of ten Brassica seed classes characterized by high texture similarity, we develop and evaluate a convolutional neural network optimized through architectural design and hyperparameter tuning. Experimental results demonstrate that the proposed framework achieves a classification accuracy of 93%, outperforming several state-of-the-art pretrained models. These findings highlight the potential of automated end-to-end deep learning models to enhance precision agriculture, providing robust and scalable solutions for seed quality monitoring and agricultural productivity.

Sprouts are gaining popularity among consumers worldwide due to their high nutritional properties. A comparative evaluation of novel and environmentally friendly pre-sowing seed treatment techniques was conducted to enhance wheat sprout production. Pulsed electromagnetic field (PEMF), cold atmospheric plasma (CAP), and high-pressure processing (HP) at 200 and 600 MPa were applied on durum wheat seeds for 3 and 10 min. The above techniques, along with ozonation (OZ), were also applied for 3 and 10 min for the “activation” of water that was used for immersion of the wheat seeds. Seed germination percentage, root and shoot length, and seedling dry weight were the measurements for the comparative evaluation of 21 treatments of seeds growing in Petri dishes. The results indicated that CAP, PEMF, and OZ treatments had positive effects on wheat sprout production, while prolonged exposure to HP processing appeared to stress the seeds. Overall, the multiple comparisons of four processing technologies, applied by two methods and at two exposure times, could be a benchmark study for further understanding the response of seeds in pre-sowing techniques.

Bambara groundnut is an underutilized legume with significant potential for enhancing nutrition and food security. However, limited knowledge of its seed systems hinders the development of targeted interventions. This study assessed the Bambara groundnut seed systems and seed quality in Ghana. A semi-structured questionnaire was used to examine seed production, conditioning, and saving practices, while 150 farmer-saved seed samples were evaluated for quality. The findings revealed that the seed system is predominantly informal, with 99% of farmers relying on their own saved seeds, which showed poor germination and emergence. About 54.4% sourced seeds from local markets, and 60.7% recycled seeds for five years or more. Seeds were stored unshelled in polypropylene bags in Tolon and shelled in the Kintampo South and Talensi Districts. Seed selection was primarily based on size (in Tolon) and visible absence of disease symptoms (in Kintampo South and Talensi). An incidence (7.6%) of seed-borne pathogens was recorded, with Aspergillus flavus (38.8%) and A. niger (16.6%) being most prevalent. Other pathogens included Macrophomina phaseolina (11.5%), Rhizopus spp. (6.5%), Curvularia lunata (5.3%), and A. fumigatus (1.9%). This study highlights the need to support community-based seed systems to improve farmers’ access to quality Bambara groundnut seed.