Current trends will secure traceability's place within the social infrastructure as a means of ensuring food safety and consumer confidence. Safety can be proven by scientific measurements but consumer confidence must be secured through the credibility of information sources. Therefore, it is important that all involved in the supply of a food product comply with industry regulations. A traceability system for securing food safety and consumer confidence from farm to table needs to cover the entire supply chain, and has to be seamless and integrated from a consumer's viewpoint. Trackable and retrospective information must be separated into two; a safety confirmation history for the product and details of dealings or movement within the distribution channel. Of these two kinds of information, what can be standardized is the distribution channel history. To this end it is necessary to promote industry-standard coding systems, access methods, and so on. In the era of globabilization and consequent increases of economic efficiency, "Food Safety" must be the key management target for food-related companies in order to secure food safety and consumer confidence. In each stage: production, processing, distribution and sale, an interlinked supply chain needs to be constituted, requiring business process reengineering.

View full abstract

The Taguchi Method has been recognized as a leading tool in quality engineering. The Taguchi Method utilizes the S/N ratio to represent the factorial effects of control variables against outputs and also takes noise factors into consideration. The Taguchi Method states that control variables with strong cross interactions should not be included in design parameters. In implementing the Taguchi Method it is necessary to select control variables without strong cross interactions. Experimentally identifying such unwanted variables is impractical due to the enormous number of combinations of control variables that must be examined. Recently, Artificial Neural Networks (ANN) have been utilized to describe various non-linear systems. The performance of combine harvesters, including their threshing function, can be considered as a typical nonlinear agricultural machinery system. In this paper, ANN were used to model these non-linear combine harvester systems in order to reveal hidden cross interactions between control variables. After selecting the proper control variables, a simulation was run based on an L18 matrix. The factorial effects of the threshing function control variables were obtained from the simulation results. Obtaining the factorial effects of the control variables facilitated design of the combine harvester threshing function. Combining the Taguchi Method and ANN was proved to be effective.

View full abstract

This paper shows that the nonlinear relationships between the personal parameters (age, gender experiences, physical conditions etc.) and the outputs of POMS (Profile of Moods State) score could be identified by using a neural network. The neural network model can noninvasively measure the POMS scores needed in order to develop a psychological response feedback control system for stress management. While the research subjects were engaged in horticultural activity, their age, gender, activity time, interest in gardening and the contents of their activities were examined in relation to the change in their POMS scores in a psychological test. The results suggested that POMS change could be investigated through stress index parameters.
The neural network model developed was tested with 10 trial subjects. The error obtained was less than 10%. The systems approach discussed in the research can be applied in the field of horticultural therapy and nonlinear modeling tools such as neural networks can identify the nonlinear systems involved in such systems.

View full abstract

This paper proposes a system architecture for a stress relief program for improving nursing home residents' (including dementia patients') quality of life through the use of horticultural activity. The system consists of three sub-systems (Reference Identifier, Program Optimizer and Outcome Evaluator). The first sub-system determines a standard stress level expressed by physiological parameters based on individual attributes such as gender, age and preferences. Those who have psychological barrier tend to avoid participation in group activity. The quality of life of residents will be improved if their barrier is eliminated or reduced. A properly designed horticultural therapy program reduces the stresses of residents. Such stress relief helps residents to overcome any psychological barrier.
The systems approach was applied to identify the management program for nursing home residents' quality of life. Neural networks were successfully utilized to model the relationships between various parameters involved in those subsystems.

View full abstract

A field management system has been developed for large-scale rice production farms. The system efficiently manages the production management information of a large number of distributed fields. The system features a field map, with geographical coordinates, as the main user interface and management data such as operations, materials and etc. are inputted and displayed on a field basis by manipulating the field map. The fields are painted in different colors according to their attributes, which helps users to know the progress of operations at a glance. The system provides an editing function that lets a user make his or her own field map. The field map can contain more than a hundred thousand of polygons. For labor management, all staff members' labor data can be inputted by PDA devices and easily summarized to account for labor productivity, greatly reducing the time that such summarizing previously required. Meteorological data obtained through MetBroker, their temporal changes and accumulated values are made available to users. Since all these basic functions are common to management systems for other crops, the system can be applied to those crops with only a small number of modifications. The system was successfully evaluated by a large-scale agricultural production company, demonstrating that it could reduce the burden of managing the information associated with a large number of fields.

View full abstract

We developed a system for automatically monitoring sheep behavior using computer vision. We designed an algorithm for identifying sheep within images. This algorithm applied thresholding to a density histogram generated from mobile objects identified by a frame difference method. To assess the performance of the prototype software, an experiment was conducted using Suffolk sheep. In the experiment, an outdoor fenced area of about 5m x 5m was constructed. Images of the behavior of the sheep were captured from a height of approximately 5m at one-second intervals from sunrise to sunset. The images were taken on both fine and cloudy days. From the captured images, sheep position coordinates were automatically measured and recorded with the prototype software. To examine the effect of the changing solar illumination, position error was calculated for each image sampled at five-minute intervals. A position error of less than 1/4 the width of the sheep occurred at a rate of 75.6% on fine days and 95.2% on cloudy days. To examine the effect of rapid changes in brightness due to cloud movement, we calculated the position error at one-second intervals within 20 seconds of the appearance or disappearance of shadows. Under these conditions, position error less than 1/4 the width of the sheep occurred at a rate of 73.7% when the sheep passed near the water tub and the feeding trough. In other cases it occurred at a rate of 90% or more.

View full abstract

Maps used in the precision farming include soil parameters related to moisture, pH, electric conductivity, soil fertility, inorganic nitrogen, NH4 and yields. We used the kriging method on a geostatistical basis to convert soil sampling data from paddy field and farms into two-dimensional maps containing position information. In this study, we developed a method to interpolate the selected soil sampling data into that needed to make a map by solving the problem of two-dimensional potential through an application of the finite element method. Soil fertility mapped by FEM, and kriging (exponential, gaussian and spherical models) were investigated. There were slight differences among the soil fertility maps in terms of frequency distribution and statistics. The average maps derived from FEM for each 10 meter-sized mesh were found to be similar to those derived by kriging.

View full abstract