Search

Vocalizations of livestock convey information about the health and behavior of the animals, and vocal analysis could be a useful method to monitor livestock. We propose a deep learning classification of vocal recordings of laying hens and cattle with the aim of automatically classifying laying hen and cattle sounds in South Korea using a deep learning model. Audio and video recordings of laying hens and cattle were acquired. We classified laying hens’ sounds into eight classes and cattle sounds into nine classes. Classified audio files were used for the development of convolutional neural network (CNN) models. Two types of CNN structures, one based on 2D ConVnet and the other based on a 1D model with long short-term memory, were developed and tested for modeling to classify the vocalizations of laying hens and cattle. The classification model based on 2D ConVnet performed better with a satisfactory classification accuracy of 75.78% for laying hens and 91.02% for cattle. Based on the results for the developed CNN models, it is expected that real-time voice monitoring could be applicable for providing animal physiological information to growers.

The sensor node is a basic unit of a greenhouse control system that collects environment data and sends them to an environment controller. It is important for the sensor node and the controller to share a common communication protocol to enable effective communication. The objective of this research was to develop a wireless sensor node complying with a communication interface standard between a sensor node and a controller and to evaluate variation in data transmission rate with distance. The TTAK.KO-06.0288-Part1/R1 standard was selected as a communication interface standard; it was analyzed and implemented with a C++ library. A sensor node and a controller were designed to communicate with the library wirelessly using Bluetooth. The system was tested in a greenhouse to evaluate the variation in data transmission rate with changes in the distance between the components. A C++ library, libgnode, was developed and successfully implemented for transmitting data from the sensor node to the controller, complying with the TTA standards. The data transmission rate was 100% up to a distance of 25 m between the sensor node and the controller. The libgnode library can be utilized to improve the interoperability between components of a greenhouse control system. However, further research into the use of long-distance wireless communication methods (e.g., Zigbee or LoRa) is warranted in order to extend the communication coverage area in the greenhouse.

th , 2016; Revised: May 20 th , 2016; Accepted: May 27 th , 2016 Purpose: In-situ water quality monitoring based on ion-selective electrodes (ISEs) is a promising technique because ISEs can be used directly in the medium to be tested, have a compact size, and are inexpensive. However, signal drift can be a major concern with on-line management systems because continuous immersion of the ISEs in water causes electrode degradation, affecting the stability, repeatability, and selectivity over time. In this study, a computer-based nitrate monitoring system including automatic electrode rinsing and calibration was developed to measure the nitrate concentration in water samples in real-time. Methods: The capabilities of two different types of poly(vinyl chloride) membrane-based ISEs, an electrode with a liquid filling and a carbon paste-based solid state electrode, were used in the monitoring system and evaluated on their sensitivities, selectivities, and durabilities. A feasibility test for the continuous detection of nitrate ions in water using the developed system was conducted using water samples obtained from various water sources. Results: Both prepared ISEs were capable of detecting low concentrations of nitrate in solution, i.e., 0.7 mg/L NO3-N. Furthermore, the electrodes have the same order of selectivity for nitrate: NO3 >> HCO3 > Cl > H2PO4 > SO4 2, and maintain their sensitivity by > 40 mV/decade over a period of 90 days. Conclusions: The use of an automated ISE-based nitrate measurement system that includes automatic electrode rinsing and two-point normalization proved to be feasible in measuring NO3-N in water samples obtained from different water sources. A one-to-one relationship between the levels of NO3-N measured with the ISEs and standard analytical instruments was obtained.

Purpose: Machine vision-based image processing methods can be useful for estimating the fresh weight of plants. This study analyzes the ability of two different image processing methods, i.e., morphological and pixel-value analysis methods, to measure the fresh weight of lettuce grown in a closed hydroponic system. Methods: Polynomial calibration models are developed to relate the number of pixels in images of leaf areas determined by the image processing methods to actual fresh weights of lettuce measured with a digital scale. The study analyzes the ability of the machine vision- based calibration models to predict the fresh weights of lettuce. Results: The coefficients of determination (> 0.93) and standard error of prediction (SEP) values (< 5 g) generated by the two developed models imply that the image processing methods could accurately estimate the fresh weight of each lettuce plant during its growing stage. Conclusions: The results demonstrate that the growing status of a lettuce plant can be estimated using leaf images and regression equations. This shows that a machine vision system installed on a plant growing bed can potentially be used to determine optimal harvest timings for efficient plant growth management.