Elsevier

Journal of Food Engineering

Volume 106, Issue 2, September 2011, Pages 170-175
Journal of Food Engineering

Research note
Evaluation of Photoshop software potential for food colorimetry

https://doi.org/10.1016/j.jfoodeng.2011.02.034Get rights and content

Abstract

This paper presents a simple method that shows how digital imaging and software analysis can be combined for color measurement. The results showed that L, a and b values from Hunter colorimeter and the digital imaging method had appropriate correlation with R2 of greater than 0.98, however color values obtained from digital imaging method can be used only to monitor the trend of color changes and relative comparison and there is a noticeable difference between L, a and b from digital imaging and values of Hunterlab colorimeter. By using equations presented in this paper, L, a and b values obtained from digital imaging method can be successfully converted to Hunterlab color values. Changes in color of Mazafati date during ripening by hot acetic acid solution were monitored by use of this method. L, a, and b values of the samples decreased during ripening.

Introduction

Color is one of main parameters that consumers evaluate and use as an indicator for the acceptance or rejection of foods (Fernandez et al., 2005, Iqbal et al., 2010, Mendoza et al., 2006, Valadez-Blanco et al., 2007). Evaluation of food color can take place by the human visual system or color measuring instruments. Visual systems involve comparison with colored references under controlled illumination. In this method in order to carry out a more objective color analysis, color standards are often used as reference materials. Unfortunately, it implies a slower inspection and requires more specialized training of the observers. For these reasons it is recommendable to determine color through the use of color measuring instrumentation. Colorimeters such as: Minolta chroma meter, Hunterlab colorimeter and Dr. Lange colorimeters are some of the most used instruments for color measurement (Leon et al., 2006). In recent years Digital imaging has been used in the food industry for quality evaluation, detection of defects, identification, grading and sorting of fruits and vegetables, meat, fish, bakery products and prepared goods (Fernandez et al., 2005, Antonelli et al., 2004, Du and Sun, 2005, Mery and Pedreschi, 2005, Zhou et al., 2004, Hatcher et al., 2004, Tanska et al., 2005, Pedreschi et al., 2006).

To measure color of different materials, various color spaces have been reported. Two frequently used color spaces are RGB and CIE Lab. RGB (Red, Green and Blue) color space consists of a three-dimensional rectangular coordinate system with R, G and B axes. A color image is represented in RGB format with these three components per pixel in the range 0–255 that their intensity electronically combined to produce a digital color picture. The available hardwares for color image processing, such as color sensors, monitors and digital cameras are geared to RGB color space. The CIE standard Lab color spaces experiments have revealed that the L, a and b values are easier related to the perceived color than other existing scales. In 1976, the Commission Internationale d’Eclairage recommended the CIE Lab or Lab, color scale for use. L is the luminance or lightness component, which ranges from 0 to 100 and parameters a (from green to red) and b (from blue to yellow) are the two chromatic components, which have no numerical limits. The Lab values are often used in research studies on food and almost always the color of foods has been measured in Lab mode (Technical services department Hunter associated laboratory, 1996, Technical services department Hunter associates laboratory Inc., 2001, Adobe Systems, 2004, Valous et al., 2009, Vaughn et al., 2009, Zhang et al., 1998).

Food color information could be accomplished using commercial software packages such as Photoshop (Adobe Systems Inc., San Jose, United States) or MATLAB (The Mathworks Inc., MA, United States). Photoshop manually can select the region of interest of product images so can evaluate the distribution of color values across a specific product or sample. Application of Photoshop software for color measurement has been previously reported in the literature (Yam and Papadakis, 2004). In this paper the potential of using digital and Photoshop software for measurement of Mazafati date fruit color change during accelerated ripening is evaluated and compared to the Hunter Lab system. For this reason the accuracy and repeatability of this method at different Lab values are investigated.

Section snippets

Material and methods

The methodology used was consisted of four steps:

  • I.

    Color measurement ability: In this step, the correlation between Hunter color values and Photoshop color values used was determined. To assess the method performance in different ranges of L, a and b, three groups of colored cards with different values of Lab were designed by Photoshop software. In each group a color parameter was considered as a variable while the other two were kept constant (for instance L was variable and a and b

Results and discussion

The correlation coefficient between Photoshop and Hunterlab was 0.991, 0.966 and 0.987 for L, a and b values, respectively (Fig. 3, Fig. 4, Fig. 5). These results show that Photoshop software can be used for determination of color parameters at a wide range of L, a, b. In spite of strong correlation (Table 3, Table 4), there are noticeable differences between Hunterlab Lab and Photoshop values and the values are not equal, therefore it seems necessary to find an appropriate way for

Conclusions

In this study the color parameters Lab were determined by use of an acquisition image chamber under controlled conditions (illumination, distance between camera and color cards, camera angle and light source) and colored cards. Evaluations of Hunter lab and Photoshop Lab values showed that Photoshop data can be used only for relative comparison of food color, however, by modification of this data with equations suggested in this paper it can be used for determination of absolute color

Acknowledgements

The authors would like to thank Mr. Mohsen Radi and Ms. Sahar-Sadat Moosavi-Nasab for their technical supports.

References (27)

  • D. Mery et al.

    Segmentation of colour food images using a robust algorithm

    Journal of Food Engineering

    (2005)
  • F. Pedreschi et al.

    Development of a computer vision system to measure the colour of potato chips

    Food Research International

    (2006)
  • S. Romani et al.

    Image characterization of potato chip appearance during frying

    Journal of Food Engineering

    (2009)
  • Cited by (145)

    • Food porn: A review from social, cultural, and economic perspectives

      2023, Critical Social Challenges in the Tourism Industry: Labor, Commodification, and Drugs
    View all citing articles on Scopus
    View full text