Development of a low cost open-source ultrasonic device for plant height measurements

Plant height is commonly used to characterize crops in various domains such as plant breeding or precision farming. Despite significant advances in sensing technologies, plant height is still very often measured manually in many research applications where high-throughput alternatives are not always...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Germain Montazeaud, Christophe Langrume, Simon Moinard, Cédric Goby, Arnaud Ducanchez, Bruno Tisseyre, Guilhem Brunel
Formato: article
Lenguaje:EN
Publicado: Elsevier 2021
Materias:
Acceso en línea:https://doaj.org/article/6c5e5d215fbb465c8df4b0ed7c696687
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Sumario:Plant height is commonly used to characterize crops in various domains such as plant breeding or precision farming. Despite significant advances in sensing technologies, plant height is still very often measured manually in many research applications where high-throughput alternatives are not always suitable. Here, we have developed a low cost, open-source ultrasonic device for semi-automated plant height measurements in small- to medium-scale applications. The main innovation compared to previous developments is the combination of a low-cost ultrasonic sensor and a plastic backscatter plate to improve plant tip detection. We compared the device to the manual method under controlled laboratory conditions and in a field experiment with 26 sorghum (Sorghum bicolor) inbred lines. The accuracy of the device was close to 1 cm in controlled conditions and 2 cm in the field, and the bias was close to 0 in both cases. In the field, measurements were 42% faster when compared with the classical ruler, and sensor-based values were strongly correlated with ruler-based values (R² = 0.9965). Overall, the device allows significant time savings while maintaining very high accuracy compared to the manual method. Its low cost (∼75 €) and compact design make it suitable for a wide range of applications, either for crop species, natural species, or model organisms. We encourage such implementations by providing all code and materials in free and open access.