Quality evaluation of Hanyuan Zanthoxylum bungeanum Maxim. Using computer vision system combined with artificial neural network: A novel method

A novel technology based on computer vision system (CVS) and artificial neural network (ANN) was developed for the quality evaluation of Hanyuan Zanthoxylum bungeanum Maxim (HZB). The quality evaluation of HZB mainly depended on its colour, odour substances, and impurities. In this study, the contents of volatile oil (VOC), total alkylamides (TALC) and impurities (IMC) were determined and used as indices for quality control of HZB. Furthermore, CVS was also performed to determine the colour parameters (RGB values) and further transforms to CIE L*, a^*, and b*. Then, ANN was carried out to analyse the correlations between colour values obtained by CVS and quality parameters of HZB (VOC, TALC, and IMC). Higher performance and stability were presented by using CVS for determining the coloristic values of HZB. In addition, the present results also showed that the established method based on ANN could be used to predict the VOC, TALC, and IMC of HZB with the R² values of 0.9991, 0.9995, and 0.9998, respectively. This novel technology based on CVS combined with ANN could be used for the rapid, non-destructive, and effective evaluation of the quality of HZB.     

Predicting the colour properties of viscose knitted fabrics using soft computing approaches

The aim of this paper was to predict the colour strength of viscose knitted fabrics by using fuzzy logic (FL) model based on dye concentration, salt concentration and alkali concentration as input variables. Moreover, the performance of fuzzy logic (FL) model is compared with that of artificial neural network (ANN) model. In addition, same parameters and data have been used in ANN model. From the experimental study, it was found that dye concentration has the main and greatest effects on the colour strength of viscose knitted fabrics. The coefficient of determination (R²), root mean square (RMS) and mean absolute errors (MAE) between the experimental colour strength and that predicted by FL model are found to be 0.977, 1.025 and 4.61%, respectively. Further, the coefficient of determination (R²), root mean square (RMS) and mean absolute errors (MAE) between the experimental colour strength and that predicted by ANN model are found to be 0.992, 0.726 and 3.28%, respectively. It was found that both ANN and FL models have ability and accuracy to predict the fabric colour strength effectively in non-linear domain. However, ANN prediction model shows higher prediction accuracy than that of Fuzzy model.     

Computer vision‐based analysis of foods: A non‐destructive colour measurement tool to monitor quality and safety

Computer vision‐based image analysis has been widely used in food industry to monitor food quality. It allows low‐cost and non‐contact measurements of colour to be performed. In this paper, two computer vision‐based image analysis approaches are discussed to extract mean colour or featured colour information from the digital images of foods. These types of information may be of particular importance as colour indicates certain chemical changes or physical properties in foods. As exemplified here, the mean CIE a^* value or browning ratio determined by means of computer vision‐based image analysis algorithms can be correlated with acrylamide content of potato chips or cookies. Or, porosity index as an important physical property of breadcrumb can be calculated easily. In this respect, computer vision‐based image analysis provides a useful tool for automatic inspection of food products in a manufacturing line, and it can be actively involved in the decision‐making process where rapid quality/safety evaluation is needed. © 2013 Society of Chemical Industry     

Computer Vision System Applied to Classification of “Manila” Mangoes During Ripening Process

Mango is an important crop that is marketed on a large scale around the world. The degree of ripeness of mangoes is an important quality attribute that has traditionally been evaluated manually through their physicochemical properties and color parameters, but recent non-destructive technologies such as computer vision systems (CVS) are emerging to replace these destructive, slow, and costly methods by others that are faster and more reliable. In the present work, physicochemical properties and color parameters obtained using a CVS at laboratory level were linked to establish the ripening stages of mango cv. “Manila.” Classification process involving multivariate analysis was applied with the aim of using only color parameters to estimate levels of ripeness. A set of 117 mangoes was used to estimate the ripening index (RPI) from the physicochemical properties, and another set of 39 mangoes was used to validate the classification process in mangoes harvest in a different season. The RPI was useful for establishing three phases of maturation, namely: pre-climacteric, climacteric, and senescence. These showed correspondences with the color changes evaluated in two color spaces (CIELAB and HSB). Principal component analysis was efficient in selecting the most significant variables and separating the mangoes into the three ripening stages. Multivariate discriminant analysis made it possible to obtain classification rates of 90 % by using only a*, b*, H and S color coordinates, the CIELAB system being, in general, more efficient at classification than HSB. The results obtained showed that CVS developed for the study can be used as a useful non-invasive, efficient method for the evaluation of the ripeness of mangoes.     

ANN-based method for olive Ripening Index automatic prediction

The influence of ripening degree of drupes during the harvesting period is well established in olive oil sector. A range of methods for expressing the stage of maturity of olives have been proposed in scientific literature. One of the most commonly adopted methods provides the evaluation of a Ripening Index (RI) on the basis of olive skin and pulp colour. Unfortunately, the RI evaluation technique is time-consuming, subjective (depending on expert skill) and depends on environmental conditions that may affect colour appearance of olives. This work describes a novel method for rapid, automatic and objective prediction of the Ripening Index of an olive lot. The method integrates a Machine Vision system, capable of performing a colour-based raw prediction of RI, with an Artificial Neural Network (ANN) based algorithm to refine it. Such a refinement is based on a set of chemical parameters (oil content, sugar content and phenol content) which are provided as input to the ANN and which can be obtained by historical curves for the region where the RI needs to be predicted. Experimental results demonstrate the effectiveness of the proposed approach.     

利用分形理论和模拟技术求解非织造布孔隙分形维数的新方法

利用数字图像技术获得非织造布表面形态的二值图像,根据分形理论用S ierp insk i地毯法求得孔隙的面积分形维数Df;用计算机模拟纵向毛细管通道,并用计盒法求得弯曲分形维数DT。孔隙分形维数的求解为研究非织造布的结构和性能提供了有效而可靠的方法。     

Identification of Seeds by Colour Imaging: Comparison of Discriminant Analysis and Artificial Neural Network

This study describes the use of colour image analysis to identify four seed varieties. A wide range of kernel measurements was obtained from digitised colour images of whole seed samples of rumex, wild oat, lucerne and vetch. The combination size, shape (including kernel seven invariant moments) and texture parameters is the major element in this investigation. Two pattern recognition approaches were attempted in the classification: stepwise discriminant analysis, which is part of statistical pattern recognition techniques, and artificial neural network. The artificial neural network was found to outperform discriminant analysis. With only three inputs, a simple three-layer perception network exhibited performances exceeding 99% both in learning and test sets. It is shown that a mixture of features improved classification from 92% for size and shape parameters to 99% for size, shape and texture parameters. Two species, totally overlapped in the morphometrical space, were well separated by texture. The best characteristics are extracted from the red channel images. Limitations of neural computing concepts are discussed with respect to seed classification.     

Edge Detection Features to Evaluate Hardness of Dates Using Monochrome Images

Date is an important fruit in the regular diets of many peoples in the Arab countries and several other parts of the world. Hardness is one of the important attributes in determining the quality of dates. Hard dates are tough, difficult to chew, unsuitable for several product preparation and ultimately fetching low market price. In general, hard dates have strong curvy and zigzag textured skin. In this study, the efficiency of edge detection features in classifying dates based on hardness using monochrome images was determined. Date samples (Fard variety) were obtained from three major dates growing regions in Oman, and classified into three grades (soft, semi-hard and hard) by a group of trained graders followed with a confirmation by an experienced grader in a commercial dates company. Individual dates were imaged using a monochrome camera (600 dates per grade; total?=?1,800 images). A total of 36 features were extracted (28 in spatial domain and 8 in frequency domain) using edge detection methods. An artificial neural network (ANN) was used to classify the dates based on hardness. The overall classification accuracies were 75 % and 87 % while using single ANN (irrespective of regions) for three-class (soft, semi-hard and hard) and two-class (soft and hard (semi-hard and hard together)) models, respectively. While using separate ANN for each region in the three-class model, the mean classification accuracies were 94 %, 59 % and 84 % for soft, semi-hard and hard dates, respectively. Similarly, for the two-class ANN model for each region, the accuracies were 95 % and 77 % for soft and hard dates, respectively. Edge detection features have a great potential in determining several surface qualities of food and agricultural products, where similar gray or color values but varying texture are found.     

How different is human hair? A critical appraisal of the reported differences in global hair fibre characteristics and properties towards defining a more relevant framework for hair type classification

This review critically appraises the reported differences in human hair fibre within three related domains of research: hair classification approaches, fibre characteristics and properties. The most common hair classification approach is based on geo-racial origin, defining three main groups: African, Asian and Caucasian hair. This classification does not account sufficiently for the worldwide hair diversity and intergroups variability in curl, shape, size and colour. A global classification into eight curl types has been proposed but may be too complex for reproducibility. Beyond that, hair cross-sectional shape and area have been found to have an inverse relation to curl: straighter fibres are circular with larger cross-sectional area, whilst the curlier fibres are elliptical with smaller cross-sectional area. These geometrical differences have been associated with bilateral vs homogenous distribution of cortical cell in curly vs straight hair respectively. However, there is no sufficient data demonstrating significant differences in hair amino composition, but proteomic studies are reporting associations of some proteins with curly hair. Eumelanin’s relative abundance has been reported in all hair colours except for red hair which has a high pheomelanin content. Higher tensile and fatigue strength of straight hair are reported, however, curly hair fragility is attributed to knotting, and crack and flow formations rather than the structural variations. African hair has been found to have the highest level of lipids, whilst the water sorption of Caucasian hair is the highest, and that of Asian hair the lowest. Not all comparative studies clearly report their hair sampling approaches. Therefore, to strengthen the robustness of comparative studies and to facilitate cross-study data comparisons, it is recommended that the following hair defining characteristics are reported in studies: hair cross sectional diameter/area, curl type, hair assembly colour, as well as where possible donor data (age/gender) and sample pooling approach.     

Recent developments in the applications of image processing techniques for food quality evaluation

Image processing techniques have been applied increasingly for food quality evaluation in recent years. This paper reviews recent advances in image processing techniques for food quality evaluation, which include charge coupled device camera, ultrasound, magnetic resonance imaging, computed tomography, and electrical tomography for image acquisition; pixel and local pre-processing approaches for image pre-processing; thresholding-based, gradient-based, region-based, and classification-based methods for image segmentation; size, shape, colour, and texture features for object measurement; and statistical, fuzzy logic, and neural network methods for classification. The promise of image processing techniques for food quality evaluation is demonstrated, and some issues which need to be resolved or investigated further to expedite the application of image processing technologies for food quality evaluation are also discussed.     

Prediction of beef eating quality from colour, marbling and wavelet texture features

Beef longissimus dorsi colour, marbling fat and surface texture are long established properties that are used in some countries by expert graders to classify beef carcasses, with subjective and inconsistent decision. As a computer vision system can deliver objective and consistent decisions rapidly and is capable of handling a greater variety of image features, attempts have been made to develop computerised predictions of eating quality based on these and other properties but have failed to adequately model the variation in eating quality. Therefore, in this study, examination of the ribeye at high magnification and consideration of a broad range of colour and marbling fat features was used to attempt to provide better information on beef eating quality. Wavelets were used to describe the image texture of the beef surface at high magnification rather than classical methods such as run lengths, difference histograms and co-occurrence matrices. Sensory panel and Instron analyses were performed on duplicate steaks to measure the quality of the beef. Using the classical statistical method of partial least squares regression (PLSR) it was possible to model a very high proportion of the variation in eating quality (r² = 0.88 for sensory overall acceptability and r² = 0.85 for 7-day WBS). Addition of non-linear texture terms to the models gave some improvements.     

A comparison of artificial neural networks and partial least squares modelling for the rapid detection of the microbial spoilage of beef fillets based on Fourier transform infrared spectral fingerprints

A series of partial least squares (PLS) models were employed to correlate spectral data from FTIR analysis with beef fillet spoilage during aerobic storage at different temperatures (0, 5, 10, 15, and 20 °C) using the dataset presented by Argyri et al. (2010). The performance of the PLS models was compared with a three-layer feed-forward artificial neural network (ANN) developed using the same dataset. FTIR spectra were collected from the surface of meat samples in parallel with microbiological analyses to enumerate total viable counts. Sensory evaluation was based on a three-point hedonic scale classifying meat samples as fresh, semi-fresh, and spoiled. The purpose of the modelling approach employed in this work was to classify beef samples in the respective quality class as well as to predict their total viable counts directly from FTIR spectra. The results obtained demonstrated that both approaches showed good performance in discriminating meat samples in one of the three predefined sensory classes. The PLS classification models showed performances ranging from 72.0 to 98.2% using the training dataset, and from 63.1 to 94.7% using independent testing dataset. The ANN classification model performed equally well in discriminating meat samples, with correct classification rates from 98.2 to 100% and 63.1 to 73.7% in the train and test sessions, respectively. PLS and ANN approaches were also applied to create models for the prediction of microbial counts. The performance of these was based on graphical plots and statistical indices (bias factor, accuracy factor, root mean square error). Furthermore, results demonstrated reasonably good correlation of total viable counts on meat surface with FTIR spectral data with PLS models presenting better performance indices compared to ANN.     

Unsound kernel identification using linear colour charge‐coupled device

Wheat is one of the most consumed grains in the world. The identification of wheat based on surface characteristics is important for the market. This study is aimed at identifying unsound kernels (Triticum durum Desf), including 710 black germ kernels, 627 broken kernels and 1169 sound kernels from several seed distributors in China. The system is mainly composed of a liner charge‐coupled device for image capture and a software package for extracting various morphological, colour and texture features. The models built by partial least squares discriminate analysis, support vector machine discrimination analysis (SVMDA) and principal component analysis‐artificial neural networks for identifying the unsound kernels have been explored. After comparisons of these three methods, it has been found that SVMDA got the best accuracy: 95.1%, 96.0% and 98.3% (black germ kernels, broken kernels and sound kernels). Obviously, the experimental results have shown that SVMDA is the most feasible and effective choice for the identification.     

Evaluation of the effect of moisture content on cereal grains by digital image analysis

Physical appearance and kernel morphology significantly affect the grade of a harvested crop in addition to other factors such as test weight, percentage of foreign matter and constituent components. Moisture content of grain can potentially affect the physical appearance and kernel morphology. The objective of this study was to evaluate the effect of moisture content on the classification capability of colour, morphology and textural features of imaged grains. Colour images of individual kernels and bulk samples of three grain types, namely Canada Western Amber Durum (CWAD) wheat, Canada Western Red Spring (CWRS) wheat and barley were acquired using a machine vision system. The grain kernels were conditioned to 12%, 14%, 16%, 18% and 20% moisture contents before imaging. Previously developed algorithms were used to extract 123 colour, 56 textural features from bulk sample images and 123 colour, 56 textural, 51 morphological features from individual kernel images. The extracted features were analysed for the effect of moisture content. Statistical classifiers and a back propagation neural network model were used for classifying the grain bulk at different moisture contents. The colour and textural features of bulk grain images were affected by the moisture content more than that of the single kernel images.     

Classification of pre-sliced pork and Turkey ham qualities based on image colour and textural features and their relationships with consumer responses

Images of three qualities of pre-sliced pork and Turkey hams were evaluated for colour and textural features to characterize and classify them, and to model the ham appearance grading and preference responses of a group of consumers. A total of 26 colour features and 40 textural features were extracted for analysis. Using Mahalanobis distance and feature inter-correlation analyses, two best colour [mean of S (saturation in HSV colour space), std. deviation of b∗, which indicates blue to yellow in L∗a∗b∗ colour space] and three textural features [entropy of b∗, contrast of H (hue of HSV colour space), entropy of R (red of RGB colour space)] for pork, and three colour (mean of R, mean of H, std. deviation of a∗, which indicates green to red in L∗a∗b∗ colour space) and two textural features [contrast of B, contrast of L∗ (luminance or lightness in L∗a∗b∗ colour space)] for Turkey hams were selected as features with the highest discriminant power. High classification performances were reached for both types of hams (>99.5% for pork and >90.5% for Turkey) using the best selected features or combinations of them. In spite of the poor/fair agreement among ham consumers as determined by Kappa analysis (Kappa-value < 0.4) for sensory grading (surface colour, colour uniformity, bitonality, texture appearance and acceptability), a dichotomous logistic regression model using the best image features was able to explain the variability of consumers’ responses for all sensorial attributes with accuracies higher than 74.1% for pork hams and 83.3% for Turkey hams.     

Prediction of beef eating quality from colour,marbling and wavelet texture features

Beef longissimus dorsi colour, marbling fat and surface texture are long established properties that are used in some countries by expert graders to classify beef carcasses, with subjective and inconsistent decision. As a computer vision system can deliver objective and consistent decisions rapidly and is capable of handling a greater variety of image features, attempts have been made to develop computerised predictions of eating quality based on these and other properties but have failed to adequately model the variation in eating quality. Therefore, in this study, examination of the ribeye at high magnification and consideration of a broad range of colour and marbling fat features was used to attempt to provide better information on beef eating quality. Wavelets were used to describe the image texture of the beef surface at high magnification rather than classical methods such as run lengths, difference histograms and co-occurrence matrices. Sensory panel and Instron analyses were performed on duplicate steaks to measure the quality of the beef. Using the classical statistical method of partial least squares regression (PLSR) it was possible to model a very high proportion of the variation in eating quality (r² = 0.88 for sensory overall acceptability and r² = 0.85 for 7-day WBS). Addition of non-linear texture terms to the models gave some improvements.     

Supervised neural network classification of pre-sliced cooked pork ham images using quaternionic singular values

The quaternionic singular value decomposition is a technique to decompose a quaternion matrix (representation of a colour image) into quaternion singular vector and singular value component matrices exposing useful properties. The objective of this study was to use a small portion of uncorrelated singular values, as robust features for the classification of sliced pork ham images, using a supervised artificial neural network classifier. Images were acquired from four qualities of sliced cooked pork ham typically consumed in Ireland (90 slices per quality), having similar appearances. Mahalanobis distances and Pearson product moment correlations were used for feature selection. Six highly discriminating features were used as input to train the neural network. An adaptive feedforward multilayer perceptron classifier was employed to obtain a suitable mapping from the input dataset. The overall correct classification performance for the training, validation and test set were 90.3%, 94.4%, and 86.1%, respectively. The results confirm that the classification performance was satisfactory. Extracting the most informative features led to the recognition of a set of different but visually quite similar textural patterns based on quaternionic singular values.     

Comparative analysis of techno-functional properties,starch digestion and protein quality of pigmented chickpea flours

The aims of this investigation were to evaluate physicochemical, functional, pasting, and thermal properties, as well as the starch and protein digestibilities of whole flours obtained from ten chickpea cultivars differing in seed coat colour (black, brown, green, red and cream). The coloured chickpeas flours contained higher amounts of bioactive compounds as total phenolics (TPC, 241.25–444.41 μg gallic acid equivalents per g), β-glucans (1.02–2.42 g/100 g), resistant starch (22.68–37.52% of total starch) and higher protein digestibility corrected amino acid scores (PDCAAS, 0.61–0.82) compared with the cream-coloured chickpea Blanco Sinaloa (C.BS). The principal component analysis showed several differences among the chemical compositions, starch digestions and seed protein qualities; in the same sense we found a correlation between TPC and starch content with their thermal properties and starch digestion. Subsequently, pigmented chickpea cultivars have potential as functional ingredients for food product development.     

Determination of cleaning end of dairy protein fouling using an online system combining ultrasonic and classification methods

Fouling and cleaning of heat exchangers in food industry are severe and costly issues and of high importance. In this study, a planar heat exchanger was constructed to produce and clean milk protein fouling similar to industry. Using a combination of an ultrasonic measuring method and classification machines cleaning should be monitored online to adapt cleaning time. After reproducible fouling deposit was built, cleaning started which was monitored using an ultrasonic measuring unit. The measured ultrasonic signal was analyzed for seven acoustic features and fed together with temperature and mass flow rate (both measured) into a classification method for decision of fouling presence or absence. For classification, artificial neural network (ANN) and support vector machine (SVM) was applied displaying detection accuracies of more than 80 % (ANN) and 94 % (SVM), respectively. Besides, the slope change of the seven acoustic features was monitored with time resulting in a cleaning time of at least 21?±?4 min. The cleaning time determined by the new sensor system is comparable with previously determined cleaning times for this setup. This study demonstrated that ultrasound based sensor systems offer a new tool to determine presence or absence of fouling and to monitor cleaning processes in the food industry with high accuracy.