Identifying Potato Varieties Using Machine Vision and Artificial Neural Networks

The objective of this study is to develop a method for identifying and discriminating 10 potato varieties by combining machine vision and artificial neural network methods. The potato varieties include Agria, Savalan, Florida, Fontaneh, Natasha, Verona, Karso, Elody, Satina, and Emrad. A total number of 72 characteristic parameters specifying color, textural, and morphological features are found among these varieties. By using principal component analysis, 16 principal features are selected for identifying and discriminating potato varieties. The data obtained from image processing were classified using linear discriminant analysis and non-linear artificial neural network method. The accuracy of discriminant analysis were 73.3, 93.3, 73.3, 40, 73.3, 73.3, 66.7, 80, 40, and 53.3%, respectively, for the varieties used in this study. The classification accuracy was improved by 100% for all the varieties using neural network analysis and the correct classification ratio was 100% using this method. It is revealed from the results that machine vision technique and neural network analysis could identify potato varieties with acceptable accuracy.     

The Prediction of Yarn Tensile Properties by Using Artificial Neural Networks

This paper presents an artificial-neural-net model for predicting yarn tensile properties. A single hidden-layer neural network trained by using the back-propagation algorithm performs a functional mapping between material properties, process variables, and the resulting yarn tensile properties. The material and process variables, namely, yarn count, blend, and front- and back-nozzle pressures on an air-jet spinning machine, arc correlated with the experimentally determined yarn properties: breaking load and breaking elongation. The neural net was trained and then used to predict the tensile properties of yarns. The errors of prediction were low despite the availability of only a relatively small data set for training, and in each ease the prediction error was less than the standard deviation of experimentation. Use of the cross-validation technique ensured that the neural net obtained a generalized mapping of the inputs and outputs.     

Modeling Thermal Conductivity of Iranian Flat Bread Using Artificial Neural Networks

The objective of this work was to develop Artificial Neural Network (ANN) based thermal conductivity (K) prediction model for Iranian flat breads. Experimental data needed for ANN models were obtained from a pilot-scale set-up. Breads were made from three different cultivars of wheat and were baked in an eclectic oven at three different baking temperatures (232°C, 249°C and 260°C). A data set of 205 conditions was used for developing ANN and empirical models. To model K using ANN, 16 different MLP (multilayer perceptron) configurations ranging from one to two hidden layers of neurons were investigated and their prediction performances were evaluated. The (4-3-5-1)-MLP network, that is a network having two hidden layers, with three neurons in its first hidden layer and five neurons in its second hidden layer, had the best results in predicting the thermal conductivity of flat bread. For this network, R², MRE, MAE and SE were 0.988, 0.6323, 1.66×10^{? 3}, and 8.56×10^?4, respectively. Overall, ANN models (with R² ≥ 0.95) performed superior than the empirical model (with R² = 0. 870).     

Ascorbyl Palmitate Efficacy in Enhancing the Accelerated Storage Stability of Canola Oil

The effect of various levels of ascorbyl palmitate (AP) and of butylated hydroxyanisole/toluene (BHA/BHT) on the accelerated storage stability of canola oils was determined by sensory, gas liquid chromatographic (GLC) and chemical evaluations. In Schaal oven tests (65°C, 0–16 days), chemical, GLC and trained sensory panel data indicated that 200 ppm AP retarded autoxidation in stored canola oils. Monoglyceride citrate (MGC) addition to oils containing 200 ppm AP did not enhance oil stability. Fluorescent light tests (7500 lux, 24°C, 0–24 hr) showed that 200 ppm AP, with or without MGC, had a limited effect in protecting canola oil from photooxidation. BHA/ BHT, at 100 ppm each, with MGC, did not improve canola oil stability.     

Oxidation on the Stability of Canola Oil Blended with Stinging Nettle Oil at Frying Temperature

The present study was carried out to investigate changes in fatty acids composition and their ratio by mixing of canola oil (CLO) with cold pressed stinging nettle oil (SNO) during heating at 180°C for 10 h, and analyzed by gas chromatography-mass spectrometry. The blended oils were prepared by gravimetrically mixing of CLO and SNO at various ratios of 60:40 and 40:60 w/w, respectively. Trans fatty acid in fresh CLO and SNO of 1.1 and 0.05%, whereas fresh blended oils of CLO:SNO (40:60, 60:40 w/w) contained 0.44, 0.91% and heated oils of CLO, SNO and CLO: SNO (40:60, 60:40 w/w) were found from 1.35 ?2.49, 0.06 ?0.09, and 0.46 ?0.51, 1.02 ?1.27%, correspondingly. The linolenic fatty acid is more prone to oxidation in heated oils and their percentage in fresh CLO, SNO and blended (40:60, 60:40) CLO:SNO samples of 10.58, 0.00, and 3.94, 6.64%, respectively. The linolenic acid was decreased from 10.27 ?6.54, 0.00, and 3.93 ?3.79%, 6.47 ?5.68% in heated CLO, SNO and CLO:SNO (40:60, 60:40 w/w) oils, respectively. The oxidation parameters were also analyzed such as free fatty acids, peroxide value, iodine value, conjugated diene and triene using standard methods. The best results of free fatty acids, peroxide value, iodine value, conjugated diene and triene were obtained in blended CLO:SNO (40:60 w/w) oil at 0.31%, 0.63 meqO₂/Kg, 0.79 g/100 g, 60 and 31.36% during 10 h. The obtained results show that mixing of CLO with SNO increased the stability against oxidation and consequently enhanced the worth of CLO during heating/frying route.     

人工神经网络在纺织工业中的应用

人工神经网络 (ａｒｔｉｆｉｃｉａｌｎｅｕｒａｌｎｅｔｗｏｒｋ ,ＡＮＮ)在纺织工业中的应用可概括为以下几类 :与纺织品的物理与机械性能有关的特性的预测 ;材料中疵点的识别、分类与分析 ;过程优化与在线监控 ;营销、计划以及其他与工业相关的工作。纺织材料的特性预测有多种不同的建模方法 ,如数学建模、经验建模、计算机模拟建模和人工神经网络建模。在这些建模方法中 ,人们发现人工神经网络是预测材料特性较为有效的方法 ,而且发现从深刻理解问题本质而发展起来的一个好的参数模型通常具有较好的性能。因此 ,神经网络最好用于那些对…     

Chemometric Amylose Modeling and Sample Selection for Global Calibration Using Artificial Neural Networks

Chemometric arnylose modeling for global calibration, using whole grain near infrared transmittance spectra andsample selection, was used in an artificial neural network (ANN), to assess the global and local models generated, based onsamples of newly bred Indica, Japonica and rice. Global sample sets had a wide range of sample variation for amylose content(0 to 25.9%). The local sample set, Japonica sample, had relatively low amylose content and a narrow sample variation(amylose; 12.3% to 21.0%). For sample selection the CENTER algorithm was applied to generate calibration, validation andstop sample sets. Spectral preprocessing was found to reduce the optimum number of partial least squares (PLS) componentsfor amylose content and thus enhance the robustness of the local calibration. The best model was found to be an ANN globalcalibration with spectral preprocessing; the next was a PLS global calibration using standard spectra. These results pose thequestion whether an ANN algorithm with spectral preprocessing could be developed for global and local calibration models orwhether PLS without spectral preprocessing should be developed for global calibration models. We suggest that global calibra-tion models incorporating an ANN may be used as a universal calibration model.     

Estimation of Dielectric Properties of Cakes Based on Porosity, Moisture Content, and Formulations Using Statistical Methods and Artificial Neural Networks

Dielectric constant (DC) and dielectric loss factor (DLF) are the two principal parameters that determine the coupling and distribution of electromagnetic energy during radiofrequency and microwave processing. In this study, chemometric methods [classical least square (CLS), principle component regression (PCR), partial least square (PLS), and artificial neural networks (ANN)] were investigated for estimation of DC and DLF values of cakes by using porosity, moisture content and main formulation components, fat content, emulsifier type (Purawave™, Lecigran™), and fat replacer type (maltodextrin, Simplesse). Chemometric methods were calibrated firstly using training data set, and then they were tested using test data set to determine estimation capability of the method. Although statistical methods (CLS, PCR and PLS) were not successful for estimation of DC and DLF values, ANN estimated the dielectric properties accurately (R ², 0.940 for DC and 0.953 for DLF). The variation of DC and DLF of the cakes when the porosity value, moisture content, and formulation components were changed were also visualized using the data predicted by trained network.     

Modelling the growth of Leuconostoc mesenteroides by Artificial Neural Networks

The combined effect of temperature (10.5 to 24.5 degrees C), pH level (5.5 to 7.5), sodium chloride level (0.25% to 6.25%) and sodium nitrite level (0 to 200 ppm) on the predicted specific growth rate (Gr), lag-time (Lag) and maximum population density (yEnd) of Leuconostoc mesenteroides under aerobic and anaerobic conditions, was studied using an Artificial Neural Network-based model (ANN) in comparison with Response Surface Methodology (RS). For both aerobic and anaerobic conditions, two types of ANN model were elaborated, unidimensional for each of the growth parameters, and multidimensional in which the three parameters Gr, Lag, and yEnd are combined. Although in general no significant statistical differences were observed between both types of model, we opted for the unidimensional model, because it obtained the lowest mean value for the standard error of prediction for generalisation. The ANN models developed provided reliable estimates for the three kinetic parameters studied; the SEP values in aerobic conditions ranged from between 2.82% for Gr, 6.05% for Lag and 10% for yEnd, a higher degree accuracy than those of the RS model (Gr: 9.54%; Lag: 8.89%; yEnd: 10.27%). Similar results were observed for anaerobic conditions. During external validation, a higher degree of accuracy (Af) and bias (Bf) were observed for the ANN model compared with the RS model. ANN predictive growth models are a valuable tool, enabling swift determination of L. mesenteroides growth parameters.     

造纸过程定量水分神经网络建模和控制

根据纸机定量水分过程的机理,提出了一种新的神经网络建模方法.在此基础上,针对该过程的大滞后、大惯性、强非线性、时变以及多变量耦合的特点,提出了一种包括定量控制和水分控制的多变量神经网络控制策略.利用该控制策略建立的自适应神经网络控制器,可以使成纸的定量和水分维持在设定的范围之内.进一步分析可以得到,该控制器是一个状态和控制作用均可跟踪的伺服系统.实际应用效果非常好.     

人工神经网络技术与纺纱质量预报

评述了人工神经网络和纺纱过程的特点 ,提出人工神经网络在纺纱质量预报中的工作原理和网络构建方法 ,并提供了国内外的应用实例和网络的实现方法 ,指出人工神经网络技术在纺纱质量预报中的广泛应用前景     

Modeling and Optimization of Electrodialytic Desalination of Fish Sauce Using Artificial Neural Networks and Genetic Algorithm

Electrodialysis (ED) has been proposed as a means to reduce sodium ion concentration in fish sauce. However, no information is so far available on the optimum condition to operate the ED process. Artificial neural network (ANN)-based models were therefore developed to predict the ED performance and changes in selected quality attributes of ED-treated fish sauce; optimum operating condition of the process was then determined via multi-objective optimization using genetic algorithm (MOGA). The optimal ANN models were able to predict the ED performance with R ²?=?0.995, fish sauce basic characteristics with R ²?=?0.992, and the concentrations of total aroma compounds and total amino acids, flavor difference, and saltiness of the treated fish sauce with R ²?=?0.999. Through the use of MOGA, the optimum condition of the ED process was the use of an applied voltage of 6.3 V and the maintenance of the residual salt concentration of the treated fish sauce of 14.3 % (w/w).     

Classification of Chinese Vinegars Using Optimized Artificial Neural Networks by Genetic Algorithm and Other Discriminant Techniques

The aims of this study were to explore the most important volatile aroma compounds of Chinese vinegars and to apply the artificial neural networks (ANN) to classify Chinese vinegars. A total of 101 volatile aroma components, which include 21 esters, 16 aldehydes, 15 acids, 19 alcohols, 10 ketones, 9 phenols, 5 pyrazines, 3 furans, and 3 miscellaneous compounds, were identified by gas chromatography mass spectrometry. On the basis of sensitivity analysis, 6 and 11 volatile aroma compounds were selected and proved to be useful for classifying Chinese vinegars by fermentation method and geographic region, respectively. The variables with the greatest contribution in the classification of Chinese vinegars by geographic region were 2-methoxy-4-methylphenol and acetic acid, whereas 3-methylbutanoic acid and furfural played the most important roles in fermentation method classification. ANN could classify Chinese vinegars based on fermentation method and geographic region with a prediction success rate of 100%. This level was higher than the accuracy of cluster analysis, linear discriminant analysis, and K-nearest neighbor. Results showed that ANN was a useful model for classifying Chinese vinegars.     

Stability and Antioxidant Activity of Beta Carotene in Conventional and High Oleic Canola Oil

Conventional canola oil (CO) and high oleic canola oil (HOCO) were stored under autoxidative and photooxidative conditions, β-carotene was added at 0 to 120 ppm. The oils were diluted in mobile phase and injected onto an HPLC column to track β-carotene changes over time. Peroxide values were followed to assess the oxidative stability of the oils. β-carotene was more stable in HOCO than in CO under autoxidative conditions, but no difference between oils was observed under photooxidative conditions. The HOCO was more stable against autoxidation, and CO was more stable against photo-oxidation. Antioxidant activity was shown by β-carotene in both oils, which contained natural tocopherols, during both autoxidative and photooxidative conditions.     

Application of Artificial Neural Networks to Predict the Oxidation of Menhaden Fish Oil Obtained from Fourier Transform Infrared Spectroscopy Method

Oxidation is a major cause of deterioration in fish oil, leading to considerable losses of quality and nutritional value. To date, the available methods to monitor lipid oxidation in foods are based on chemical analysis. Fourier transform infrared spectroscopy (FTIR) is an alternative technique for the study of molecular structure and compositional changes in a wide range of foods. The objectives of this study were to use attenuated total reflectance-FTIR for evaluating oxidative quality and application of artificial neural network analysis (ANN), a mathematical model, to predict the oxidative values of Menhaden fish oil. The oil was stored in the presence of light at room temperature. The oxidation was measured for primary and secondary oxidative change; peroxide value (PV) and anisidine value (AnV), respectively, using FTIR were compared with chemical analysis each day during the 3 weeks of storage. The wavenumber and absorbance values of FTIR spectra were applied to predict the oxidative values of the oil by using ANN. Inputs consisted of wavenumber and absorbance outputs were composed of PV and AnV. It was found that changes in the region between 3,500 and 1,700 cm^-1 and absorbance were related to PV and AnV of the chemical analysis (R ² > 0.85). FTIR spectroscopy with the aid of ANN demonstrates its potential as an alternative and rapid technique rather than a conventional method for prediction of food lipids oxidation.     

大蒜油提取工艺优化及神经网络模型的研究

采用溶剂法对大蒜油的提取进行了研究 ,并通过正交试验近似获得了较优的工艺参数组合 :温度 80℃ ,料液比 1∶7、时间 2h ,得率 0 2 5 8%。在此基础上 ,利用神经网络建立了工艺参数和得率之间关系的数学模型 ,为工艺人员了解不同工艺条件下的大蒜油得率提供了一个有效的计算工具。