Nondestructive measurement of husk-covered corn kernel layer dynamic moisture content in the field

Highlights

• A noninvasive method for measuring husk-covered corn kernel moisture is proposed.

• Dynamic moisture of corn kernel during filling stage is in-situ obtained in the field.

• The moisture of corn kernel follows exponential decay functions before maturity.

Abstract

Corn kernel moisture content is the most frequently used indicator for kernel maturity prediction and determination of the appropriate mechanized harvesting time. Corn varieties with lower moisture contents and thus faster dehydration of the corn kernels after physiological maturity are suitable for high plant density and mechanized harvesting. However, nondestructive measurement of the dynamic moisture content of the husk-covered corn kernel layer before physiological maturity in the field remains challenging at present. Therefore, in this study, a nondestructive measurement method based on the vector reflection coefficient measurement technique is proposed and developed for in situ monitoring of the husk-covered corn kernel layer moisture in the field. Based on the one-port calibration procedure used for the handheld moisture detection device, quadratic regression models were established for the corn kernel layer moisture using impedance parameters from 51 husk-covered corn samples. Results showed that this method enabled nondestructive estimation of the husk-covered corn kernel layer moisture ($R^2=0.62$) with a 3.53% root mean square error ($\mathit{RMSE}$) using a serial capacitor $C_s$ and the husk moisture ($R^2=0.85$) with a $RMSE$ of 7.12%. Additionally, intact corn ears were measured in the field from the dent stage to the maturity stage using this device to investigate the dynamic moisture contents of the kernels and husks. Results showed that the measured moisture content of kernels before physiological maturity followed exponential decay functions with increasing numbers of days, and the mean difference between the measured kernel moisture contents of the intact corn ears and the oven-drying moisture contents of these kernels is 3.72%. In conclusion, the handheld moisture detection device allows us to measure the moisture of husk-covered corn kernel layers nondestructively before maturity in the field. This study thus provides a new way to obtain corn kernel dehydration curves in situ before maturity and estimate the corn maturity time.

Introduction

Rapid dehydration of corn kernels is an important factor in corn breeding and cultivation (Zhou et al., 2018), particularly for high plant densities and when mechanized harvesting is used. In mechanized harvesting of corn, if the kernel moisture content is too high, it generally results in mechanical damage; however, if the kernel moisture content is too low, it can cause harvesting losses. Nondestructive measurement of corn kernel moisture content and monitoring of the kernel dehydration process during the grain filling stage in the field to predict the kernel maturity time accurately are thus essential processes to maximize the corn yield (Herrmann et al., 2005, Martinez-Feria et al., 2019). The kernel moisture content at maturity affects the moisture content of the harvested kernels. In fact, the dehydration of corn kernels in the field includes two phases: i) maturation dehydration, and ii) post-maturity dehydration. Before the kernels reach physiological maturity, i.e., the maturation dehydration stage, large amounts of relocated assimilates are displaced into the kernels (Maiorano et al., 2014). The kernel dehydration rate in this phase is regulated by the growth and development processes. The husk moisture has a positive effect on kernel dehydration, and lower husk moisture content results in lower moisture in the harvested kernels (Kang and Zuber, 1989). At the post-maturity dehydration phase, moisture exchange between the kernels and the atmosphere is the only dissipation route available for kernel moisture loss and the kernel dehydration rate is not only determined by the air temperature and the relative humidity (Maiorano et al., 2014) but also is affected by the number, thickness and weight of the husks (Zhou et al., 2016).

The conventional method used to perform grain moisture measurements is the oven drying method. However, this method is time-consuming and involves some tedious procedures that cannot be adapted for nondestructive in situ measurement of the dynamic moisture content of husk-covered corn kernels in the field. Benefiting from the continuing development of microwave and radio frequency technologies, rapid nondestructive methods for measurement of grain moisture based on the dielectric properties of agricultural materials have been widely applied. Several studies measuring in situ corn kernel moisture content have been performed using pin-type or ring-type grain moisture analyzers (Fan et al., 2020, Filipenco et al., 2013, Hurburgh, 1985, Kang and Zuber, 1989, Kang et al., 1978, Reid et al., 2014, Reid et al., 2010, Spry, 1990, Spry, 1992, Zhang et al., 2016). After the kernels reach physiological maturity, the physiological role of the corn husks has been completed and the husks can then be removed to accelerate the kernel dehydration process. At this stage, the kernel moisture content can be measured using the pin-type or ring-type devices (Fan et al., 2020, Zhang et al., 2016). However, before the kernels reach physiological maturity, the husks protect the kernels and also serve as important nutrient assimilation and storage organs for the developing kernels (Araus et al., 2012, Mesterházy et al., 2012, Ning et al., 2012, Pengelly et al., 2011, Yan et al., 2011). Therefore, before maturity, nondestructive measurement of the moisture content of husk-covered corn kernels is a prerequisite. Although we previously developed a ring-type sensor connected to a network analyzer for post-maturity kernel moisture detection in corn without husks, it is not currently possible to determine in situ the moisture content of husk-covered kernels directly and noninvasively (Zhang et al., 2016). Other researchers have used pin-type sensors to impale husk-covered corn kernels to measure their moisture contents (Filipenco et al., 2013, Reid et al., 2014, Reid et al., 2010). However, these pin-type devices damage the developing kernels, particularly in the early stages of corn ear growth, which can cause disease and infection (Kebebe et al., 2015, Reid et al., 2014). In addition, the husks and the corn cobs affect the repeatability of kernel moisture content measurements when using these pin-type devices. Challenges thus remain in nondestructive measurement of the dynamic moisture content of husk-covered corn kernels in the field at present.

To measure the dynamic moisture content of husk-covered corn kernels noninvasively in the field during the grain filling stage, we propose a new measurement method based on vector reflection coefficients and a handheld device, and use quadratic function models for fitting of the corn moisture content. First, after the one-port calibration procedure of the handheld moisture detection device is performed, the vector reflection coefficients of the husk-covered corn ears, which are coupled via the double-ring electrodes, are measured to extract the impedance parameters. The quadratic function models that relate the impedance parameters and the moisture contents of the kernels and husks are then established. Finally, based on these models, the dynamic moisture contents of husk-covered corn kernels in the field are measured from the corn dent stage to the maturity stage. Using this handheld device, the dynamic moisture content of kernels in husk-covered corn ears before physiological maturity can be measured nondestructively during different stages of kernel development in the field and the dehydration curve of the corn kernels can thus be obtained. Therefore, the proposed method can help us to determine the appropriate mechanized harvesting time and to screen the corn varieties that are suitable for high plant density and mechanized harvesting.