My team and I at Washington State University developed a robust and affordable thermal imaging system optimized for agricultural applications. The system can be used for continuous monitoring of crops in the field at a fixed position or aboard moving irrigation systems (e.g. center pivot). The imager combines thermal and RGB imagery and can pull data from a microclimate unit wirelessly. In combination with appropriate models and algorithms, the thermal-RGB imaging system can be used for creating real-time evapotranspiration and prescription maps, and irrigation scheduling.
In this post I will discuss this thermal-RGB imaging system.
Development
The key features of this thermal-RGB imaging system (Osroosh et al., 2018) are summarized below:
Robust and inexpensive ($400 for the prototype)
Relies on Raspberry Pi single-board computer as core
Uses FLIR Lepton thermal module & Raspberry Pi RGB camera module
Has dedicated on-board GUI & interval timer shooting
IoT and communication capabilities inherited from Raspberry Pi
A precise timer turns imager on/off at specified times of day (saves power)
Weather proof 3D designed & printed enclosure
Raw images are saved in binary & JPG format
The field model of the imaging system can be used to create a wireless network for high resolution spatial and temporal monitoring of agricultural fields and orchards. The design of the imager allows for creating a star network of imaging units in the field to obtain real-time surface temperature data from plant canopies. A power management panel was specifically designed for the imagers to turn them on/off at specified times of day. This feature would save power and allow for unattended continuous monitoring in the field for a long period of time.
​Besides the field imager, our team developed a functioning handheld thermal-multispectral imager prototype based on the same concept. The idea was to have a handheld camera for convenient field monitoring of plant water status and health on an as-needed basis. The key differences with the thermal-RGB imager are that the handheld camera is equipped with a touchscreen, and is self-powered using battery, and can be used by personnel with somewhat less technical skills. The multispectral images allow for the calculation of NDVI. Please check out BINA Pro for the advanced model of this imager with on-board image processing capability.
Image Processing Algorithm
​Our team also developed a computer vision algorithm to extract the surface temperatures of specific target (e.g. leaves, fruits, animals) and canopy cover percentage from captured images. The algorithm uses RGB images to identify desired target (e.g. leaves, fruit) in the thermal image and separate it from the background. It then calculates the average temperature of all the remaining image pixels (i.e. target).
​
Mounting Imagers on a Center Pivot Irrigation System
In the spring of 2017, we mounted two of these thermal-RGB imagers on a center pivot irrigation system in a mint field near Toppenish, WA. The center pivot was retrofitted with Medium Elevation Spray Application (MESA) and Low Elevation Spray Application (LESA) systems. The computer vision algorithm was used here to extract sunlit leaf temperatures and canopy cover percentage from images. The algorithm removed the soil background, and calculated the surface temperature of sunlit leaves and canopy coverage. The system survived the harsh weather of central Washington during an entire growing season.
Thermal-RGB imagers mounted on center pivot irrigation system
Apple and Cherry Fruit Surface Temperature Monitoring
​In another experiment, our team investigated the potential of this custom-built thermal-RGB imagery system for monitoring fruit surface temperature. The results of this research showed that the inexpensive imager could replace an extensive (Price: ~$40,000) high-resolution thermal camera in detecting critical surface temperatures.
In another experiment, our team determined the feasibility of the thermal-RGB imager for detecting cherry fruit surface wetness level and duration. We carried out an experiment in plots of cherry varieties at the Roza Farm of Washington State University (Osroosh and Peters, 2019). We used a rain simulator to wet cherries. The in-field sensing setup included two custom-built thermal-RGB imagers, a microclimate-measuring unit and two leaf wetness sensors. The computer vision algorithm identified leaves and cherries in thermal and RGB images and extracted the surface temperatures. By utilizing the imaging system, decision aid tools may be developed for efficient rainwater removing to prevent fruit cracking.
Citation
Ihuoma, S.O., Madramootoo, C.A., 2017. Recent advances in crop water stress detection. Comput. Electron. Agr., 141: 267–275.
Khanal, S., Fulton, J., Shearer, S., 2017. An overview of current and potential applications of thermal remote sensing in precision agriculture. Comput. Electron. Agr., 139: 22–32.
Osroosh, Y. et al., 2019. Detecting fruit surface wetness using a custom-built low-resolution thermal-RGB imager. Computers and Electronics in Agriculture, 157: 509-517.
Osroosh, Y. et al., 2018. Economical thermal-RGB imaging system for monitoring agricultural crops. Computers and Electronics in Agriculture, 147: 34–43.
Comentários