Application of thermal imaging for plant disease detection

Abstract

The effects of plant diseases on agricultural production worldwide contribute to significant economic and post-harvest losses. To maintain the sustainability of the farming sector, the early detection of plant and pathogens is essential. Non - destructive methods for tracking the health conditions of plants in real-time applications are among the most realistic and feasible in this regard. Owing to non - destructive methods and non - contact measuring devices, thermal imaging advancement has become an essential technique in all fields. Thermal imaging operates by emitting infrared irradiation in all materials. The method, therefore, uses radiation to create an image of the thermal distribution of the body surface. This approach is useful for all areas where the difference in temperature helps in the area/object analyses. Therefore, this paper briefly explores the potential for thermal imaging to be used to detect plant disease in the control area and the field.

A greenhouse‐based high‐throughput phenotyping platform for identification and genetic dissection of resistance to Aphanomyces root rot in field pea

“…Therefore, improving the phenotyping platform by reducing the focal length and building a phenotyping chamber can increase resolution and light intensity while reducing noise to improve digital indices to represent the visual scores. In addition, rapidly evolving phenotyping tools, including multispectral, hyperspectral, or thermal imaging (Hashim et al., 2020; Marzougui et al., 2019; Sankaran et al., 2019), also can be implemented along with machine learning algorithms (Mohanty et al., 2016; Shruthi et al., 2019) to further enhance the speed, precision, and accuracy of ARR phenotyping. Nevertheless, our pilot study provided valuable insights into the utility of this platform for disease evaluation and characterization both at the phenotype and DNA level.…”

A greenhouse-based high-throughput phenotyping platform for identification and genetic dissection of resistance to Aphanomyces root rot in field pea

“…Therefore, improving the phenotyping platform by reducing the focal length and building a phenotyping chamber can increase resolution and light intensity while reducing noise to improve digital indices to represent the visual scores. In addition, rapidly evolving phenotyping tools, including multispectral, hyperspectral, or thermal imaging (Marzougui et al 2019; Sankaran, Quirós, and Miklas 2019; Hashim et al 2020), can also be implemented along with machine learning algorithms (Shruthi et al 2019; Mohanty et al 2016) to further enhance the speed, precision, and accuracy of ARR phenotyping. Nevertheless, our pilot study provided valuable insights into the utility of this platform for disease evaluation and characterization both at the phenotype- and DNA-level.…”