Sensor–algorithm co-design with two-dimensional optical sensing and spatiotemporal labelling for seed passage detection in precision planters
Computers and Electronics in Agriculture, cilt.254, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 254
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.compag.2026.112278
- Dergi Adı: Computers and Electronics in Agriculture
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Applied Science & Technology Source, BIOSIS, Compendex, Environment Index, Geobase, INSPEC, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO), Technology Collection (ProQuest)
- Anahtar Kelimeler: Precision planting, Seed detection, Seed singulation, Spatiotemporal labelling algorithm, Two-dimensional optical sensor
- Akdeniz Üniversitesi Adresli: Evet
Özet
Accurate seed detection in precision planters is challenging when seeds pass through the seed tube in close spatial or temporal proximity. This work presents a two-dimensional (2D) optical sensor array combined with a spatiotemporal labelling algorithm, which is designed to overcome the limitations of conventional one-dimensional optical and impact-based sensors. The system creates a virtual sensing plane using two perpendicular optical barriers and applies connected component labelling to resolve overlapping shadow events. Experiments involving corn, soybean, and sunflower seeds were conducted under varying seed flow rates, sensor–tube distances, and tube inclinations. These experiments were validated using 500 fps high-speed video. The results demonstrate that the 2D sensing approach achieves a high level of detection accuracy, exceeding 97 % for corn and soybean seeds and 98 % for sunflower seeds when the independent-axis method is applied at short sensor distances. The independent-axis detection strategy also reduced the number of missed and extra detections associated with closely spaced seed passages compared with the combined detection method. Most undetected seeds occurred near the boundaries of the sensing area or at inter-seed intervals below 10 ms. The sensor–algorithm co-design approach provides a robust and computationally efficient alternative to camera-based systems for real-time seed monitoring.