Sign in Subscribe
Agricultural industry

Advances in Wheat Seed Quality Detection through End-to-End Decision Fusion

Strategic Insights

— 1 min read

Researchers at Fujian Agriculture and Forestry University have made significant progress in developing a novel method for rapid detection of wheat seed quality. By integrating hyperspectral imaging and computer vision techniques, the team has created an end-to-end decision fusion deep convolutional neural network (DCNN) model that can detect seed quality with high accuracy. This breakthrough has the potential to improve storage management, field performance, and flour quality.

Key Takeaways:

  • The research team employed a combination of hyperspectral imaging (HSI) and computer vision (CV) to capture surface data from both the embryo (EM) and endosperm (EN) of wheat seeds.
  • The integration of HSI and CV showed considerable promise in seed quality assessment, with a validation set accuracy of 65.1-89.2%.
  • The end-to-end decision fusion DCNN model, integrating HSI-EM, HSI-EN, CV-EM, and CV-EN data, achieved the highest accuracy in both training (94.3%) and validation (93.8%) sets.
  • The proposed model simplifies the training process compared to traditional two-stage fusion methods, making it a potentially efficient alternative for rapid, individual kernel quality detection and control during wheat production.
  • The research demonstrated that applying the proposed model to seed lot screening increased the proportion of high-quality seeds from 47.7% to 93.4%.

Statistics:

  • 1000 high-quality seeds and 1098 deteriorated seeds were analyzed in the study.
  • The validation set accuracy ranged from 65.1% to 89.2%.
  • The decision fusion-based DCNN model achieved a training accuracy of 94.3% and a validation accuracy of 93.8%.
  • The proportion of high-quality seeds increased from 47.7% to 93.4% after applying the proposed model to seed lot screening.

Sources:

  • End-to-end deep fusion of hyperspectral imaging and computer vision techniques for rapid detection of wheat seed quality. Artificial Intelligence in Agriculture, 2025,15(3):537-549.
  • NewsRx. Researchers at Fujian Agriculture and Forestry University Have Published New Study Findings on Agriculture (End-to-end deep fusion of hyperspectral imaging and computer vision techniques for rapid detection of wheat seed quality). Agriculture Week. September 11, 2025; p 377.

Read more