Integration of computer vision and electronic nose as non-destructive systems for saffron adulteration detection

Highlights

• Color and aroma are the most important organoleptic characteristic of saffron.

• An integrated system based on computer vision and electronic nose was developed.

• The developed system was successfully utilized for saffron adulteration detection.

• Aroma characteristic variables were more effective than color variables.

Abstract

This work deals with the development and evaluation of an integrated system based on computer vision system (CVS) and electronic nose (e-nose) for saffron adulteration detection. Ten saffron samples adulterated with two common illegal constituents, namely, Artificially Colored Safflower (ACS) and Artificially Colored Yellow Styles of Saffron (ACYSS) at levels ranging from 10 to 50% (w/w) were characterized in this work. First, the developed CVS and e-nose system were integrated to form a unit system. This set up was utilized to extract color and aroma characteristic variables of each sample. The extracted variables were processed using Principal Component Analysis (PCA), Hierarchical Cluster Analysis (HCA), and Support Vectors Machines (SVMs) to demonstrate the discrimination capability of the developed system. Two multilayer artificial neural network (ANN-MLP) models were also employed for saffron color and aroma strength prediction based on ISO standards. PCA and HCA results of the color and aroma datasets revealed that the adulterated samples have different color and aroma strength compared to authentic saffron and they can clearly be distinguished. SVMs classifier showed good agreement with the PCA results and reached 89% and 100% success rate in the recognition of the different saffron samples based on their color and aroma datasets, respectively. Results of the two ANN-MLP models proved that the developed system is capable of differentiating the authentic and adulterated saffron samples based on their color and aroma strength ($R_{\mathit{Color}\mathit{analysis}}^2⩾0.95$ and $R_{\mathit{Aroma}\mathit{analysis}}^2⩾0.97$).

Introduction

Saffron (Crocus sativus L.) is commercially important and widely consumed as spice. It possesses desirable flavor, therapeutic and medicinal properties. Due to the high demand for saffron, its price has been steadily increasing. Thus, saffron spice has been the subject of various adulterations, such as mixing with foreign materials to increase the volume and weight of its commercial lot. The most frequently encountered extraneous and adulterant materials are artificially colored yellow styles of saffron (ACYSS) and artificially colored safflower (Carthamus tinctorius L.) (ACS) (Heidarbeigi et al., 2014). On the other hand, saffron quality could be influenced by the geographical location of production, drying procedures, and storage conditions (Maghsoodi et al., 2012). This makes the important task of saffron quality monitoring more complicated. Saffron quality includes several main attributes such as color, aroma, and taste which are determined by its main chemical compounds, namely, Crocins (C₄₄H₆₄O₂₄), Picrocrocin (C₁₆H₂₆O₇), and Safranal (C₁₀H₁₄O), respectively (Maggi et al., 2009). Saffron samples receive different quality grades containing different amounts of these chemical compounds. Adulteration in saffron samples can reduce the amounts (per volume) of its main chemical compounds and thus its quality.

Different analytical methods for the detection of saffron quality and their adulteration exist, including Near Infrared spectroscopy (NIR) (Zalacain et al., 2004), thin layer chromatography (TLC) (Pathan et al., 2009), gas chromatography mass spectroscopy (GC–MS) (Jalali-Heravi et al., 2009), liquid chromatography mass spectroscopy (LC-MS) (Verma and Middha, 2009), UV–Vis spectroscopy (Sabatino et al., 2011), high performance liquid chromatography (HPLC) (Sheikh et al., 2013), Nuclear magnetic resonance (Petrakis et al., 2015), proton transfer reaction mass spectrometry (PTR-MS) (Nenadis et al., 2016), and Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) (Petrakis and Polissiou, 2017). These methods are accurate and sensitive, but their industrial applicability is hindered by their time consuming and off-line nature, high costs, and the need for specialist operators (Ghasemi-Varnamkhasti et al., 2012a). Advances and developments in sensor technology, chemometrics and artificial intelligence make it possible to develop instruments based on artificial senses such as electronic eye or computer vision (CVS), electronic nose (e-nose) and electronic tongue (e-tongue) systems capable of measuring and characterizing color, aroma and taste of saffron. Details about these techniques have been expounded earlier (Kiani et al., 2016a, Ghasemi-Varnamkhasti et al., 2012b, Ghasemi-Varnamkhasti et al., 2016). Both CVS and e-nose techniques require little sample preparation and allow large data sets to be acquired in a short time. These two methods are nondestructive, inexpensive, and do not require skilled operator. Relatively speaking, e-tongue is much more complex, its sensors are contact type, and samples should be in liquid form. Sometimes, saffron adulteration may be so subtle that only using a single nondestructive evaluation method (such as CVS) is not capable of detecting it. More recently, applications of nondestructive tools in food quality assessment have been documented (Men et al., 2014, Forina et al., 2015, Borras et al., 2015, Peris and Escuder-Gilabert, 2016, Kiani et al., 2016b, Khulal et al., 2017). The authors of such reports have emphasized on the fact that integration of different techniques can provide more information about various aspects of the food materials and consequently the final judgment on quality indices is more accurate and reliable. To date, no report on integration of CVS and e-nose for saffron quality characterization has been published. Thus, the objectives of this study include: (1) development of an intelligent technique based on the integration of computer vision and e-nose systems coupled with multivariate methods, (2) evaluation of the developed system for detection of adulterated saffron samples based on their color and aroma profiling.