Apple mealiness detection using fluorescence and self-organising maps

doi:10.1016/S0168-1699(03)00014-0

Computers and Electronics in Agriculture

Volume 40, Issues 1–3, October 2003, Pages 103-114

https://doi.org/10.1016/S0168-1699(03)00014-0 Get rights and content

Abstract

The chlorophyll fluorescence kinetics of ‘Jonagold’ and ‘Cox’ apples, stored under different conditions to induce mealiness, were measured. Three different storage conditions were considered causing three mealiness levels: not mealy, moderately and strongly mealy. Also destructive measurements of the texture (firmness, hardness, juice content and soluble solids content) were done. Classification into different mealiness levels based on the fluorescence measurements was more performant than a classification based on the destructive measurements. To estimate the mealiness level in a non-destructive way from the fluorescence features, a number of different classifiers were constructed. Quadratic discriminants and supervised and unsupervised neural networks were tested and compared. The self-organising map gives promising results when compared with the multi-layer perceptrons and quadratic discriminant analysis. The different advantages of the constructed classifiers suggest that fluorescence can be used in an automatic sorting line to assess certain types of mealiness.

Introduction

With the newest sorting machines, it is possible to separate apples according to its colour, size and clearly visible damages like hail damage, scab and older bruises, simply and solely based on imaging with colour cameras. However, these advanced machines are not able to select on internal quality attributes (like firmness, sugar content, ripeness). Texture of the flesh of the fruit is considered as one of the most important quality attributes, besides appearance and flavour, selected by consumers (Szczesniak and Kahn, 1971, Szczesniak and Kleyn, 1963). One of the undesirable texture changes is the development of a mealy texture, which results in dry and soft fruit. There have been proposed a lot of measurement techniques for a non-destructive assessment of texture in literature (Steinmetz et al., 1996), but none of them has been applied yet on sorting machines as a common tool for distinction of different texture levels. One of the promising techniques is chlorophyll fluorescence proposed by Song et al. (1997) for assessing the senescence and the accompanying decrease in firmness of apples.

Generally, chlorophyll fluorescence is used to investigate the photosynthetic activity of leaves and to detect a wide variety of stresses in plants. A lot of factors, directly or indirectly acting on the photosynthesis, will influence the chlorophyll fluorescence: light intensity, temperature, humidity, gas composition, senescence, herbicide treatment, species, drought, nutrient shortages, diseases, the ‘entire’ prehistory of a plant. For the apple fruit, changes in fluorescence can be linked to the ageing and ripening, causing decrease of firmness, chlorophyll loss and loss of photosynthetic activity per unit chlorophyll. Combined with advanced neural network (NN)-based techniques like the self-organising map (SOM), an automated high-performance classifier based on fluorescence kinetics response has been constructed.

Section snippets

Mealiness in apples

Mealiness causes a quality devaluation of the fresh fruit and reduces its consumption. A market research carried out in Spain interviewing retailers showed some interesting results. Retailers were perfectly able to distinguish the mealy product from a fresh product, identifying it correctly with terms as: non-juicy, soft, etc. However, a lot of different terms were used to describe mealiness. The negative influence of mealiness on the consumption was very clear: sales decrease and prices go

Fluorescence data

In Fig. 1, the mean fluorescence curves for different mealiness levels for Jonagold and Cox are presented. At three points of each curve, the standard deviation around the mean is given. The fluorescence values decreased with mealiness level for the two cultivars. The fluorescence was higher for Cox than for Jonagold at the three mealiness levels. Also, the red side of the apples had a slight lower fluorescence than the green side. This difference lowered when the apples became mealier. There

Discussion

This work addressed the non-destructive estimation of mealiness induced by pectin degradation associated with senescence. The procedure that was used to produce the samples for the experiments followed the protocol developed in the FAIR project CT95-0302: “Mealiness in fruits: consumer perception and means for detection”.

A difference in the general fluorescence level was found for the two varieties of apples, Jonagold and Cox. As a consequence, differences for other varieties can be expected.

Conclusions

Chlorophyll fluorescence can be an attribute to an advanced quality sorting of apple fruit due to its speed and non-destructive character, but there still remain some problems to be solved. There was an effort to link the chlorophyll fluorescence with the mealiness of apples, supposing mealiness and chlorophyll degradation go hand in hand, but also senescence causes chlorophyll degradation and changes in the chlorophyll fluorescence. The influence of ageing is not taken into account in this

References (13)

Barreiro, P., Ortiz, C., Ruiz-Altisent, M., Recasens, I., Asensio, M.A., 1998. Identification of mealy apples by...
P. Barreiro et al.
Comparison between sensory and instrumental measurements for mealiness assessment in apples: a collaborative test
J. Texture Stud.
(1998)
C.M. Bishop
Neural Networks for Pattern Recognition
(1995)
De Smedt, V., 2000. Measurement and modelling of mealiness in apples. Ph.D. Thesis. K.U. Leuven,...
D.V. Fisher
Mealiness and quality of Delicious apples as affected by growing conditions, maturity and storage techniques
Sci. Agric.
(1943)
R.A. Johnson et al.
Applied Multivariate Statistical Analysis
(1998)

There are more references available in the full text version of this article.

Cited by (28)

Analyses of microstructure and cell wall polysaccharides of flesh tissues provide insights into cultivar difference in mealy patterns developed in apple fruit
2020, Food Chemistry
Citation Excerpt :
Secondly, a moderate mealiness pattern occurred in ‘Red Delicious’ apple after 14 days of storage. The period can be prolonged with the ripening temperature decreased, as demonstrated previously on ‘Red Delicious’ and ‘Jonagold’ apples (Arefi, Moghaddam, Hassanpour, Mollazade, & Motlagh, 2016; Moshou, Wahlen, Strasser, Schenk, & Ramon, 2003). Thirdly, a slow mealiness pattern was discovered in ‘Ralls Genet’ apple after a long-period storage of 49 days, similar to several slow-softening apple cultivars, such as ‘Scifresh’ (Ng et al., 2015), ‘Kanzi’ (Gwanpua et al., 2016) and ‘Honeycrisp’ (Harb, Gapper, Giovannoni, & Watkins, 2012).
Mealiness is one of the most important textural failure of apple fruit and four patterns of mealiness involving five apple cultivars were identified as the rapid, moderate, slow and none, requiring 3, 7, 14, 49 days at 25 °C, respectively. In comparison with the non-mealy ‘Fuji’ apple, parenchyma cells of mealy apples became detached and remained intact. Highly methyl-esterified homogalacturonan was strongly immunolabeled in the cell wall of slow and non-mealy apples. The mobility of water was enhanced in the cell wall during mealiness. Principal components analysis of FTIR spectra discriminated the cell wall materials (CWM) based on the mealiness progress. Heavy loss of CWM and its water-insoluble fractions but limited increase of water-soluble fractions, and the increase of crystalline micelles of CWM were closely associated with the mealiness progress. Overall, the occurrence of mealiness might attribute to structural, physical and biochemical modifications of CWM during tissue senescence.
Classification of 1-methylcyclopropene treated apples by fluorescence fingerprint using partial least squares discriminant analysis with stepwise selectivity ratio variable selection method
2018, Chemometrics and Intelligent Laboratory Systems
Citation Excerpt :
Fluorescence properties have been used as follows. For apples, they are used to monitor changes during maturation, ripening, and senescence [11,15], sort apples by quality [16,17], detect mealiness [18], and assess the flesh firmness of apples [11]. For apple juice, fluorescence properties are used to assess its quality during processing [19], determine the degree of naturalness [20], analyze its browning [21], and detect pesticide residues [22].
In this study, we investigated the potential of using fluorescence fingerprint (FF) for nondestructive identification of apples treated with 1-methylcyclopropene (1-MCP). In total, 442 apples of two cultivars (Fuji and Orin) and different storage times (0, 4, 5, 6, and 8 months) were assessed. The classification model used in this study was built using partial least squares discriminant analysis (PLSDA) with the stepwise selectivity ratio (SR) method. The stepwise SR method is a recursive variable selection method proposed in this study. FF was capable of classifying 1-MCP-treated apples with accuracies of 91.23%, 89.74%, and 90.17% for calibration, cross-validation, and validation results, respectively. PLSDA with the stepwise SR method could identify four aggregations of wavelength conditions, which are important to the classification. In addition, a non-targeted approach was taken to screen the metabolites characterizing 1-MCP-treated and control apples by liquid chromatography-mass spectrometry (LC/MS) and nuclear magnetic resonance (NMR) spectroscopy. The observed difference in metabolic profiles may contribute to the difference in the fluorescence profiles of 1-MCP treated and control apples.
Optical analysis using monochromatic imaging-based spatially-resolved technique capable of detecting mealiness in apple fruit
2017, Scientia Horticulturae
Citation Excerpt :
Except for being objective, instrumental tests still suffer from destructive and time-consuming disadvantages, and therefore, they are more suitable for standard tests and not for online applications. During the past years, different non-destructive testes, such as ultrasonic (Bechar et al., 2005; Mizrach et al., 2003), nuclear magnetic resonance spectroscopy/imaging (Barreiro et al., 2002; Barreiro et al., 2000), florescence spectroscopy (Moshou et al., 2003), near-infrared spectroscopy (Mehinagic et al., 2003), time-resolved spectroscopy (Valero et al., 2005), hyperspectral imaging (Huang and Lu, 2010; Huang et al., 2012), and biospeckle imaging (Arefi et al., 2016), have been investigated with the aim of mealiness detection in apple. Despite of some good results achieved by the previous studies, there is still need of investigating less expensive, rapid, and reliable techniques for detection of mealiness.
Mealy apples are often rejected by consumers due to their soft and dry tissue. As mealiness is an internal disorder without any noticeable external symptoms, it is necessary to use methods capable of non-destructive description of the tissue. This study is intended to investigate the possibility of non-destructive apple mealiness classification using spatially-resolved light scattering imaging technique. To this aim, scattering images of ‘Fuji’ apples were acquired at 650 and 980 nm. The scattering profiles were analyzed by Farrell, modified Lorentzian, modified Gompertz, and Gaussian–Lorentzian models. Statistical results showed significant changes in most of the parameters of the models while apples became mealy. Poor apple mealiness classification accuracies were achieved for PCR, PLSR, and ANN classifiers of three outputs assigned to fresh, semi-mealy, and mealy apples. The results improved when the number of classifier outputs was reduced to two classes. In that case, the best results were obtained for ANN models developed based on a combination of all effective optical parameters acquired at both wavelengths. In that case, classification accuracy of 76% and 82% was resulted in for non-mealy and mealy apples, respectively. Afterwards, classification of the non-mealy apples into fresh and semi-mealy was carried out, which resulted in 88% of fresh and 59% of semi-mealy apples to be correctly classified. Results of this study showed the possibility of application of monochromatic imaging-based spatially-resolved technique for apple mealiness classification, however, better performances would be expected in the case of increasing the number of wavelengths.
Non-destructive identification of mealy apples using biospeckle imaging
2016, Postharvest Biology and Technology
Citation Excerpt :
Therefore, it can be concluded that mealiness development during long-cold storage is inevitable. Mealiness in apple fruit is considered an unwanted parameter that significantly affects its marketability (Moshou et al., 2003). Therefore, it is necessary to monitor apples for their mealiness.
An apple that is soft and lacks juiciness during consumption is characterized as a mealy fruit. Since mealiness in apple is considered as an undesirable quality parameter, this work aimed at developing classification models based on biospeckle imaging in order to recognize mealy from non-mealy apples. To evaluate the potential of biospeckle imaging in detection of mealiness, a total of 760 ‘Red Delicious’ apples were harvested. The 540 samples were stored under cold conditions for various time periods from 0 to 5 months. The remaining samples were kept at 20 °C and 95% relative humidity for 10–26 days. Biospeckle images of each apple were acquired at wavelengths of 680 and 780 nm, respectively. Biospeckle imaging was immediately followed by a confined compression test to measure fruit stiffness and juiciness. These parameters were used to categorize the samples into three classes called fresh, semi-mealy, and mealy. Results of clustering based on a self-organizing map showed that apple fruit are mealy when their stiffness and juiciness are less than 20 kN m⁻¹ and 5 cm², respectively. After classification of apples by destructive methods, time-historical speckle patterns were configured and biospeckle features such as the inertia moment, the absolute value of differences, and autocorrelation were extracted. Results showed that biospeckle activity for fresh samples was higher than semi-mealy and mealy. Finally, several neural network models were developed to classify apples. First, classification of apples into mealy and non-mealy classes was carried out, and then non-mealy apples were classified into fresh and semi-mealy classes. The best classification accuracy for fresh (81.7%) and semi-mealy (70.9%) apples were achieved at 780 nm. However, much better result (77.3% accuracy) for classification of mealy apples was observed at the wavelength of 680 nm.
Clustering of the self-organizing map reveals profiles of farm profitability and upscaling weights
2015, Neurocomputing
Profitability and other economic aspects of farming in Finland are analyzed using clustering of the self-organizing map. The analysis of profitability bookkeeping data reveals several interesting relationships between the monitored financial variables. Economic profiles of farms are presented based on the clustering, and the findings are confirmed with statistical tests. A weight optimization system is proposed for upscaling financial figures of the sample of profitability bookkeeping farms to the whole country level. The system output is analyzed, and it is confirmed that the most important large and medium-sized enterprises are represented well by the sample. Furthermore, it seems that the utilized arable area is the key factor in guiding the weight optimization process. These findings may turn out to be useful in developing the sampling of bookkeeping farms in the future.
Texture measurement approaches in fresh and processed foods - A review
2013, Food Research International
Knowledge of textural properties is important for stakeholders in the food value chain including producers, postharvest handlers, processors, marketers and consumers. For fresh foods such as fruit and vegetable, textural properties such as firmness are widely used as indices of readiness to harvest (maturity) to meet requirements for long term handling, storage and acceptability by the consumer. For processed foods, understanding texture properties is important for the control of processing operations such as heating, frying and drying to attain desired quality attributes of the end product. Texture measurement is therefore one of the most common techniques and procedures in food and postharvest research and industrial practice. Various approaches have been used to evaluate the sensory attributes of texture in foods. However, the high cost and time consumption of organizing panelists and preparing food limit their use, and often, sensory texture evaluation is applied in combination with instrumental measurement. Objective tests using a wide range of instruments are the most widely adopted approaches to texture measurement. Texture measurement instruments range from simple hand-held devices to the Instron machine and texture analyzer which provide time-series data of product deformation thereby allowing a wide range of texture attributes to be calculated from force–time or force–displacement data. In recent times, the application of novel and emerging non-invasive technologies such as near-infrared spectroscopy and hyper-spectral imaging to measure texture attributes has increased in both fresh and processed foods. Increasing demand for rapid, cost-effective and non-invasive measurement of texture remains a challenge in the food industry. The relationships between sensory evaluation and instrumental measurement of food texture are also discussed, which shows the importance of multidisciplinary collaboration in this field.

View all citing articles on Scopus

View full text

Apple mealiness detection using fluorescence and self-organising maps

Abstract

Introduction

Section snippets

Mealiness in apples

Fluorescence data

Discussion

Conclusions

Comparison between sensory and instrumental measurements for mealiness assessment in apples: a collaborative test

J. Texture Stud.

Neural Networks for Pattern Recognition

Mealiness and quality of Delicious apples as affected by growing conditions, maturity and storage techniques

Sci. Agric.

Applied Multivariate Statistical Analysis