Investigating The Possibilities Of Document Cameras For Quality Assessment Of Foodstuffs By Measuring Of Color

Investigating the possibilities of document cameras for quality assessment of foodstuffs by measuring of color

Stanka Baycheva1, Zlatin Zlatev1, Antoaneta Dimitrova1

Trakia University – Stara Zagora

Faculty of Engineering and Technology

38 Graf Ignatiev Str., 8600 Yambol, BULGARIA

E-mail: [anonimizat]

Abstract

The possibilities for use of interactive presentation devices as powerful tools for increasing students interest, activity, motivation and participation are analyzed. Established is the possibility of measuring color with document camera in direction of determining the color characteristics of food products. The effectiveness of this type of measurement is confirmed by the study results.

Keywords: Interactive Presentation System, Interactive Whiteboard, Document camera, Color measurement

1 Introduction

Overhead projector and Diaprojectors in past years have been the main tool for visualization of learning in university audiences. Followed by multimedia projectors that with rapid pace are widely used as a tool and in modern stage whiteboards in audiences are replaced with interactive presentation systems (IPS). The interactive whiteboard (IWB) becomes significant, and in some cases mandatory element of the interactive classroom of the future. Its contents if necessary, can be seen and shared by the personal devices (tablets, laptops) of all students, and in addition can be integrated with other modern technical tools for training (Dineva&Nedeva, 2013; Shivacheva at al, 2015).

The document-camera enables the presentation in real time on items on the IWB with high quality and speed. The device can also be used to create pictures or video of real objects and text, some models are fully integrated with the software and hardware of IWB. The document camera can be used for visualization of static and dynamic objects for demonstration of experiments, increase image and presentation objects under the microscope on IWB (Pehlivanova&Ducheva, 2011; Nedeva&Dineva, 2013; Stoykova, 2014).

In recent years there has been increasing interest in using the document camera for recognition of geometric objects that are drawn on the document (triangle, circle, line), text recognition and mathematical formulas, and for objectively determining the quality characteristics of objects as food, making it a means of training courses related to technical non-destructive methods for assessing the quality of these products (Zlatev at al, 2013).

Interest in the practice represents the accuracy and realism of the images obtained with a document camera. This also applies to the identification of components of food products (such as pork, eggs, bread) which substantially have a complex structure, color, and change the surface characteristics upon storage. These characteristics of foodstuffs put out the assess of the quality of the food products with technical tools (Kirilova at al, 2010; Mladenov at al, 2014).

The aim of this report is to establish the possibility of measuring color with document camera with direction of determining the color characteristics of food products.

2 Exposure

The color of foodstuffs is difficult to assess because in different parts of the product at the same surface structure is not same. By digital image analysis of the surface color characteristics of foodstuffs is possible to be segmented images as individual specific areas and the entire surface of the studied object (Vasilev, 2016). The information gathered will establish objective criteria for assessing the quality of the food products. This type of information is not achievable with other devices.

By the colorimeter can be measured the color characteristics of object (Smits, 2010). This gives precise resultsbut they are receive point by point. Hence this type of measurement is time consuming and difficult to measure spatial data. Furthermore, the colorimeters have a significantly higher price compared with digital cameras. These advantages of cameras such as ability for contactless measuring of the color simultaneously in several areas on the object surface, aroused interest among researchers, because such devices can be used for quickly get information about the color of objects.

3 Results and discussion

To realize the objective of the study is built experimental arrangement presented in Figure 1. The experimental system consists an interactive presentation system include personal computer, interactive whiteboard, software to interact with the whiteboard. Used are two specialized document cameras Epson ELPDC11, Epson ELPDC06 and Web camera.

A comparative analysis is made of the possibility of measured values of color with these document cameras.

As reference values were used those measured with a colorimeter using a calibrated digital sensor for measuring the color in the RGB color model (Figure 2b). Used is reference with 24 colors (Figure 2a).

Color sensor is constructed as a module of the sensor for colors TCS34725 (Adafruit, 2016), which has an integrated IR filter and sensor elements for RGB color and white light. It has a digital I2C interface and features very high sensitivity and dynamic range 3800000: 1, allowing the use of the module behind the dark glass.

As a criterion for assessing the accuracy of the measurement with document camera compared with data obtained from the colorimeter were used correlation coefficient and relative error.

The correlation analysis is made in the following sequence:

Checked are the homogeneity and linearity of the data which revealed that the data are homogeneous and linear;

Determined are the correlation coefficient for the color components of the presented color models.

Figure 3 shows correlation coefficients for the R component of the RGB color model. Epson ELPDC11 and Web camera show homogeneity of the data, while Epson ELPDC06 they are not well grouped around the the linear model by which they are presented.

Table 1. Correlation between the values of the color components, as measured with a colorimeter and camera

In Table 1 are inflicted the correlation coefficients for the tested color models – RGB, HSV, YIQ, Ycbcr, for the three document cameras and correlation coefficients with values above 0,95 are marked.

The comparative analysis is made based on the relative error in measuring of color with the three document cameras with mathematical relationship [1].

Figure 4 shows graphics of the relative error ΔX, determined for the three cameras.

Table 2. Minimum and maximum values of the relative measurement error

Table 2 shows the minimum and maximum relative error in measuring color depending on the used document camera.

From the results presented it is seen that the highest indicators – high correlation with measurements with a colorimeter over 0,95 and low relative error of 2-7% have a document camera Epson ELPDC11, and with the lowest indicators is a Web camera.

The results obtained confirm reported in the literature (Zlatev at al, 2013; Stoykova, 2014)] significant number of shortcomings in the implementation of Web Cameras as document cameras for the purpose of visualization and measurement. Document cameras are designed to give a clear image to have realistic images while Web cameras are used primarily for video contacts locally or over the Internet and their manufacturers does not provide functionality for visualization and measurement.

4 Conclusion

The results of the analysis of the available literature show that in modern stage interactive systems for presentation incorporating interactive whiteboard, personal computer, special software and accessories, increase the level of visualization teaching content and increase the interest of students to the studied disciplines.

By conducted comparative analysis is found that by document cameras can be receive complete information about the measurement object at the pixel level, rather than in separate parts of the object as in the standard colorimeters.

The effectiveness of document cameras in color measurement is confirmed by the research results. The resulting correlation between the color measured with document cameras and that obtained with the reference color sensor is above 0,9.

Acknowledgements

This work was partially supported by research project granted by Faculty of “Engineering and Technology” – Yambol, Bulgaria with subject “Contactless methods for evaluation of main quality characteristics of dairy products”.

5 References

Adafruit TCS34725 (2016): https://www.adafruit.com/product/1334, accesed 2016

Dineva S., V. Nedeva (2013): How Technology Influence [anonimizat] and Success, ICVL 2013, Romania, ISSN: 1844-8933, pp.123-128

Kirilova E., P. Daskalov, R. Tsonev, Ts. Draganova (2010): Selection of colour features for recognition of Fusarium damaged corn seeds, Annual Conference RU & SU, 2010, vol.49, book 3.1, pp. 125-130, ISSN 1311-3321, (in Bulgarian)

Mladenov M. I., E. D. Dimitrov, M. P. Dejanov, S. M. Penchev (2014): Hyperspectral imaging system based on „point scan” spectrophotometer. Proceedings of International Conference "Automatics and Informatics'2014", Sofia, Bulgaria, 2014, I-39-I-42

Nedeva V., S. Dineva (2013): Design and development of efficient e-learning courses, ICVL 2013, Romania, ISSN: 1844-8933, pp.108-115

Pehlivanova, M., Zl. Ducheva (2011): Training in a virtual learning environment in the theoretical module – a factor for development of "responsible" driver, The 6th International Conference on Virtual Learning ICVL 2011, University of Bucharest and „Babeș-Bolyai“ University of ClujNapoca, Romania, ISSN: 1844-8933, pp.275-280

Smits B. (2010): An RGB to Spectrum Conversion for Reflectances, University of Utah, January 21 2000, pp.1-10

Shivacheva G., V. Nedeva, M. Yaneva, D. Georgieva (2015): Software for building virtual laboratories. XXIV International scientific conference “Management and quality” for young scientists, ISSN: 1314-4669, pp. 292-300

Stoykova V. (2014): Evaluation of the application of interactive presentation systems in higher education, ARTTE, Applied Researches in Technics, Technologies and Education, Journal of the Faculty of Technics and Technologies, Trakia University, Vol. 2, No. 3, ISSN 1314-8796, pp.286-300

Vasilev M. (2016): Image processing for color diagnosis of diseases in yellow cheese, Applied scientific journal, Innovation and entrepreneurship, vol. 4, No 1, ISSN 1314-9253, pp.25-35

Zlatev Z., K. Dobreva, V. Bochev (2013): Analysis of implementation of computer vision system in food technology education, Annual conference of University of Rousse, ISSN 1311-3321, vol.52, No.10.2, pp.229-233