IoT Enabled, Leaf Wetness Sensor on the Flexible Substrates for In-Situ Plant Disease Management

Plant diseases, weeds, and pests adversely affect crop yields. In this paper, the researchers have worked on early plant disease detection models to predict so that the spread of diseases could be prevented. These models monitor parameters like leaf wetness duration (LWD), ambient temperature, and relative humidity to detect the probability of plant disease.

Leaf wetness sensors (LWS) identify and quantify leaf wetness duration. Mechanical leaf wetness sensors could detect the presence of moisture on the leaves based on weight changes. The electronic leaf wetness sensors detect water on the leaves by measuring the sensor's capacitance or resistance variations. Generally, these sensors have interdigitated electrodes (IDE) mounted on the printed circuit board (PCB) or a ceramic plate. Electronic LWS are reliable, accurate, and precise but unaffordable to the farmers.

The researchers proposed a prototype of an Internet of Things (IoT)-enabled LWS fabricated on flexible substrates.

The LWS prototype was developed by mounting aluminium IDEs on a flexible polyimide substrate. The prototype interface consists of an LWS, a capacitance-to-frequency converter, a temperature and humidity sensor, a power management unit, a microcontroller, and a WiFi module. The sensor data from LWS was recorded and processed by an efficient, IoT-enabled, and low-power signal conditioning circuit developed in-house. A power gating scheme reduces the overall energy consumption of the system.

The researchers analyzed and calibrated the proposed LWS prototype in field experiments. As it is always challenging to calibrate the sensors during in-situ measurements, the researchers have also validated the performance with the commercially available LWS (Phytos 31: LWS-L12).

The prototype and commercial (Phytos 31) LWSs were used for a month to measure moisture in situ. About 1600 data samples were collected. The leaf wetness sensors placed on the medicinal plant Ocimum tenuiflorum (Tulsi) leaves detect and record moisture content. The sensing surface area of the prototype LWS was marked as 0%, 25%, 50%, 75%, and 100%. An LCR meter measures the sensor capacitance for each case.

The capacitance of the fabricated sensor increases monotonically with an increase in water molecule surface area. The water molecules strengthen the polarization between the IDEs, thus increasing the capacitance. The change in capacitance gives a measure of moisture on the leaf surface. An indigenous microcontroller pre-processed the sensor data of both LWSs to calculate the leaf wetness duration (LWD). The field measurement results show the discrepancy between the LWDs of the LWS prototype and Phytos 31 to be within acceptable limits.

The experiments showed only a 6% change in sensor capacitance for a temperature variation between 20°C and 65°C. The proposed LWS prototype is also sufficiently temperature-insensitive.

The proposed prototype of LWS fabricated on a flexible substrate could be a potential candidate for the in-situ measurements of LWD. The study showed that it is possible to develop an affordable, flexible, lightweight, sensitive, and temperature-insensitive plant disease detector. In the future, an IoT-enabled leaf wetness sensor fabricated on the flexible substrate could effectively and efficiently be used for the early prediction of plant diseases based on the data collected for the crop cycle.