Engineering in Agriculture, Environment and Food最新文献

Cocoons of the mulberry silkworm Bombyx mori L. are the main raw material for the silk production. Currently, at the market, their quality assessment and pricing are done on a few random samples by manual method, which is shaking cocoons with hand and assessing the generated sound, due to the absence of automated systems and time constraint. This manual method is subjective, laborious and prone to errors. A novel automated cocoon quality assessment system is proposed, which not only classifies them into good and defective ones but also subclassifies the later into dried and mute cocoons. A unique vibration impact acoustic emission (VIAE) is generated from each category due to the difference in the physical state of pupa inside the cocoon. In this system, the cocoons were vibrated using a plastic arm attached to a servo motor driven by Arduino board and the VIAE so generated was recorded by two microphones. A computer loaded with a custom-made algorithm preprocess the VIAE and compared its area under the curve of power spectral density against the pre-known threshold values, to identify the cocoon category. This automated system could successfully classify 86 cocoons with 100% accuracy in 4 s (excluding the duration of VIAE recording). This is better than the manual method in terms of accuracy, cost and skilled laborer dependency. This could make it a good replacement for the manual method to ensure the fairer cocoon trade in the market and better silk quality in the reeling centers.

This paper utilizes convolutional neural network (CNN) to identify peach leaf disease infected by Xanthomonas campestris. Transfer learning was used to fine-tune AlexNet. Feature visualization from the trained CNN indicate the excellent ability of self-learned features. Three comparative experiments were conducted to compare the performance of CNN with the traditional classification methods including Support Vector Machine (SVM), K-Nearest Neighbor (KNN) and Back Propagation (BP) neural network in identifying peach leaves. Confusion matrix of each method displays that CNN can identify the peach leaves affected by Xanthomonas campestris with the accuracy of 100%. ROC (Receiver Operating Characteristic) curves and AUC (Area Under ROC Curve) values, an overall performance measurement, show that CNN achieves higher performance with AUC value of 0.9999. The test of significant experiment shows that CNN is significantly superior to the other three mentioned methods, which the p-values is 0.0343 (vs.SVM), 0.0181 (vs.KNN) and 0.0292 (vs.BP). In a word, CNN is superior to the state-of-the-art in identifying diseased peach leaves.

To rapidly and accurately detect the quality of blueberry, hyperspectral imaging (HSI) technique was used to simultaneously detect the soluble solids content (SSC) and firmness (FI) of blueberry. In total, 204 blueberry samples, including 164 samples in Calibration set and 40 samples in prediction set, were investigated in this study. Multi-stage successive projections algorithm (MS-SPA) and SPA1/SPA2 were proposed to select a few feature wavelengths from the spectral region of 450–950 nm. Prediction models were developed based on partial least squares regression (PLSR), support vector regression (SVR) and back propagation neural network (BPNN) model. The results showed that prediction model based on MS-SPA performed better in prediction results. Furthermore, the prediction based on BPNN model was better than that based on PLSR and SVR models, which used full spectrum (FS), SPA1/SPA2, MS-SPA, respectively, to select feature wavelengths. This research suggested that MS-SPA-BPNN model, which obtained the best prediction results of SSC (R_P = 0.894, RMSEP = 0.220), and FI (R_P = 0.843, RMSE = 0.225), was a reliable tool to detect SSC and FI simultaneously. The visualization of distribution map of parameters was an intuitive and convenient measurement for quality detection of blueberry. The method could provide a theoretical basis for developing an online detecting and grading system of blueberry quality based on multispectral imaging technique.

Agriculture remains a vital sector for most countries. It presents the main source of food for the population of the world. However, it faces a big challenge: producing more and better while increasing the sustainability with a reasonable use of natural resources, reducing environmental degradation as well as adapting to climate change. Hence, it is extremely important to switch from traditional agricultural methods to modern agriculture. Smart Agriculture is one of the solutions to deal with the growing demand for food while meeting sustainability requirements. In Smart Agriculture, the role of information is increasing. Information on weather conditions, soils, diseases, insects, seeds, fertilizers, etc. constitutes an important contribution to the economic and sustainable development of this sector. Smart management consists of collecting, transmitting, selecting and analyzing data. As the amount of agricultural data increases significantly, robust analytical techniques capable of processing and analyzing large amounts of data to obtain more reliable information and much more accurate predictions are essential. Data Mining is expected to play an important role in Smart Agriculture for managing real-time data analysis with massive data. The aim of this paper is to review ongoing studies and research on smart agriculture using the recent practice of Data Mining, to solve a variety of agricultural problems.

This research represents an evaluation study of the linear and non-linear mathematical methods applied to predict the biogas flow rate in anaerobic digestion processes. The anaerobic digestion model No.1 was used to generate the process data. For the prediction of the biogas flow rate the partially least squares regression, the locally weighted regression and the artificial neural networks were used. Two metaheuristic tools, here a genetic algorithm and an ant colony optimization algorithm were applied to improve the prediction models. They carried out the variable selection procedure. The implemented mathematical models could successfully perform the prediction of the biogas flow rate. Nevertheless, more robust and accurate prediction of the biogas flow rate was done with the help of the artificial neural networks. Here the error of prediction was about 9% while the coefficient of determination reached 0.97.

In recent years, several studies have indicated that modeling techniques based on artificial intelligence can be used for efficient prediction of food industry-related variables. In this study, machine learning methods were used to predict the permeate flux of pomegranate juice in a membrane clarification system based on membrane material, pore size, pressure, flow rate, and processing time. The experimental data were modeled using curve fitting, fuzzy inference system (FIS), artificial neural networks (ANN), and adaptive neuro-fuzzy inference system (ANFIS). Results showed that the permeate flux is a function of time and a power equation can predict the permeate flux with MSE of 0.0136. FIS, ANN and ANFIS models resulted in MSEs equal to 0.0495, 0.0145, and 0.0045 for permeate flux prediction, respectively. According to these findings, ANFIS has resulted in more reliable performance which can be used as an acceptable model in the prediction of permeate flux. The optimum architecture for the ANN was obtained 5-22-1 whilst the architecture of ANFIS models for PVDF and MCE membranes were 3-7-12-12-1 and 4-9-24-24-1, respectively. The results of this study can be used to predict the amount of permeate flux in the absence of experimental data and/or for interpolation and extrapolation of the permeate flux.

Practical applications

One of the problems in juice membrane clarification is the accumulation and deposition of rejected compounds on membrane surfaces or inside its pores which results in a membrane fouling. On the other hand, several parameters can have influence on fouling and predictions of juice permeate flux during the membrane processing whereas they are important in industrial applications. Therefore, providing a model which able to predict the permeate flux having the value of effective input parameters seems to be useful. In this regard, several artificial methods can be used.

Pea (Pisum sativum L) and chickpea (Cicer arietinum L) are important grain legumes grown in the Palouse region of the Pacific Northwest United States. The USDA-ARS grain legume breeding program in this region focuses on developing pea and chickpea varieties with high yield potential, resistance to biotic and abiotic stresses, and superior agronomic characteristics. In this study, aerial high resolution multispectral imaging was evaluated to phenotype yield potential differences among genotypes in green pea, yellow pea and chickpea. Five experiments (three field pea and two chickpea) with 10–25 varieties grown at two locations (Pullman, Washington; Genesee, Idaho) were assessed. Images were acquired approximately 60, 70 and 90 days after planting (DAP) at 110 m above ground level. Normalized difference vegetation index (NDVI), green normalized difference vegetation index, soil adjusted vegetation index (SAVI) and simple ratio (SR) image based features (SUM, MIN, MAX, MEAN) were extracted. In most cases, the MEAN NDVI data was found to be consistently correlated with dry seed yield (p < 0.05), with green pea genotypes showing strongest relationship (r = 0.64–0.93 at about 70 DAP, both during “plot-by-plot” and “by genotype” comparisons). The MEAN SAVI and SR values were also strongly correlated with yield at 61–72 DAP in most of the pea experiments. The data collected during flowering and early pod development phenological growth stages was found to be useful in yield estimation. The developed methods can be used for early generation evaluation in breeding programs, where yield cannot be estimated due to limited seed availability.

A mechanized sugarcane production system with small machinery is important because it is good farming management, lowers carbon, saves energy and conserves the environment. Making a database is necessary to achieve high working efficiency and low fuel consumption of farm machines such as harvesters and tractors.

Mechanizing Okinawa farms with small machines is important. Two small sugarcane harvesters were recently introduced to Okinawa. The time and fuel consumption to operate, harvesting, hauling out, stopping, and traveling forward and backward were measured. A computer program to estimate these variables was developed based on past and current performance tests. There was little difference between estimated values and measured data.

In this approach, an amino-functionalized silica coated multiwall carbon nanotube (AminMagMWCNTs@SiO₂), for the first time, was rationally designed, prepared, and then investigated as an adsorbent for the adsorption and removal of Pb (II) and Cd (II) from environmental samples. The properties of synthesized magnetic nanoadsorbents were analyzed by Fourier transform infrared spectroscopy (FT-IR), X-ray powder diffraction (XRD), transmission electron microscopy (TEM) and scanning electron microscopy (SEM). The diameter of magnetic nanoadsorbents was in the range of 60–80 nm. The effects of various parameters on the adsorption efficiency were simultaneously studied using a chemometric design. The variables of interest were the amount of nanoadsorbent, pH and ultrasonication time. The experimental parameters were optimized using a Box–Behnken design and the response surface equations were derived. The removal of magnetic nanoadsorbents from the aqueous solution was simply achieved by applying an external magnetic field following the adsorption process. The adsorption efficiencies of AminMagMWCNTs@SiO₂ nanoadsorbent for Pb (II) and Cd (II) ions were in the range of 98–104% under the optimum condition. The results demonstrated that the amino-functionalized MagMWCNTs@SiO₂ nanoadsorbent could be used as a simple, efficient, regenerable and cost-consuming material for the removal of desired heavy metal ions from environmental water and soil samples.

Distributions of thermo-physical properties: such as porosity and thermal conductivity of Japanese apricot and pear during storage were determined based on X-ray CT image analysis. Japanese apricot was stored at 25 °C, whereas pear was stored at 25 °C and 5 °C. Average CT value was determined based on a series of X-ray CT images captured for each whole fruit. At the end of storage period, the average CT value decreased in Japanese apricot and pear at 25 °C, whereas it was the almost same as pear stored at 5 °C. Porosity increased, whereas thermal conductivity slightly decreased at 25 °C in Japanese apricot and pear. As a result of the experiment, it seemed that the internal structure of pear stored at 5 °C was well maintained during storage. Conversely, void space progressed in Japanese apricot and pear during storage at 25 °C. The porosity and thermal conductivity distributions were visualized based on the CT image during storage.