ARTTE

Applied Researches in Technics, Technologies and Education

Journal of the Faculty of Technics and Technologies, Trakia University https://sites.google.com/a/trakia-uni.bg/artte/

ASSESSMENT THE QUALITY OF JAPANESE QUAIL EGGS Julieta Ilieva, Zlatin Zlatev Trakia University, Faculty of Technics and Technologies Graf Ignatiev 38, 8602 Yambol, Bulgaria e-mail: [email protected], [email protected] Abstract: In the present study a computer vision system is used to sort Japanese quail eggs into predefined categories. The system uses the output values of the image histogram. The developed system can be a milestone for developing egg sorting machines on the basis of colour of egg shell. The accuracy of the system can reach even people without experience who can use this system to identify colour of egg shell. With the use of computer vision in the testing process instead of personal judgments, time-saving can be achieved, and the quality can also be improved in the quail farm operation. Keywords: computer vision, quail eggs, color features, histogram analysis.

1.

INTRODUCTION

The development of methodology for objective grading of eggs in the last three decades has significantly reduced the place of subjective decisions. Subjective decisions can hardly be satisfactory and sufficient criterion on a large scale for a wide range of operations. For example, there are factors such as visual fatigue, lack of color memory, differences in color differentiation between individuals, differences in lighting and others. They significantly influence the decision rights in determining the visual quality of a product. Moreover, the human eye measures only the appearance of the product and in a number of products to be found not only external but also internal defects. Visually this can be done only after the destruction of the products. That is why there is a need for new ideas, methods and algorithms to refine and fully automates the entire process associated with grading [5]. Organoleptic testing of eggs is prominently in the control system and quality control of finished products along with physico-chemical and microbiological analysis. This assessment aims to ensure consistency of the eggs with the standards in terms of organoleptic indicators and assessment of their suitability for consumption. Organoleptic tests shall be conducted by people with good visual, olfactory, taste and tactile possibilities for objective judgment. To ensure competitiveness in the sensitive quality egg the market studies are necessary to objectively measure the quality [4]. A part of the produced eggs /from 10% to 30 %/ has been damaged due to improper storage and transportation each year. That leads to the appearance and growth of microorganisms and fungi. Because of these deterioration indicators the eggs cause allergies and toxic reactions in humans and animals. The practice is essential to identify at an early stage of development, even before the visible changes have occurred that are characteristic for the development of eggs. Therefore more research is directed to a spectral analysis in which apart from the surface changes can be measured and the internal structure of the product. Much of the published results, analysis and evaluation of the eggs in the field of spectral analysis set different quality parameters such as proteins, protein content, dry matter, moisture, conductivity, acidity, different monotoksines, infections that are mainly indicators of the internal structure eggs [3, 6]. There are limited numbers of publications of assessment the characteristics that are associated with the surface texture and color characteristics of quail eggs.

ARTTE Vol. 1, No. 3-4, 2013 ISSN 1314-8788 (print), ISSN 1314-8796 (online) 181

ARTTE

Applied Researches in Technics, Technologies and Education

Journal of the Faculty of Technics and Technologies, Trakia University https://sites.google.com/a/trakia-uni.bg/artte/

The basic methods for evaluation of the quality are microscopic, biological and centrifugation. The eggs are reviewed by an expert with an eye, loupe or microscope to determine the type of deviation from the quality indicators. The methods are subjective and requires considerable time to process the results. Diagnostic accuracy is not high depending on the qualification of the expert. That is why the creation of high effective automated technology to assess the quality of the eggs is an important task [6, 9]. The concept of Hazard Analysis and Critical Control Point (HACCP) provides science-based system for food safety. Prevention is the key element of the HACCP program. In this approach it is possible to produce the safest eggs and egg products by preventing, eliminating or reducing potential biological, physical or chemical hazards for food safety, whether they are naturally in food, they are collected from the environment or generated by variation in the production process. HACCP rules and regulations are related to the manufacturing plants and are applying to all egg producers [1]. In recent years Japanese quails are bred as laboratory birds because of their easy breeding, early puberty, short intervals playback performance per eggs. Quail are popular as a source of meat and eggs [6]. Eggs of different birds have the same nutritional composition and potential for food uses. On the other hand, the information on the quality of the eggs and their use for nutritional purposes is essentially limited to chicken eggs [2]. The eggs are well studied in terms of quality but in some literature data on other species and breeds of birds. For the end user are important the external characteristics – the purity of egg freshness, the weight, the strength shell and the interior – the quality of the yolk and whites, the ratio of the components of the egg and the chemical composition [9]. These characteristics are important in industrial breeding of quails. The purpose of this paper is to develop an algorithm and a program to determine the main characteristics of Japanese quail eggs by color features. 2.

MATERIAL AND METHODS

50 eggs from Japanese quails aged 15 weeks farmed in the village of Boyanovo, Yambol region divided into cells of 1 male and 2 female are used. In the study is used computer vision system consisting of the following components [4]:  Video Camera – the camera is mounted on a movable stand that can change the height of the image shooting. The images were obtained at a height of 20cm.  Lighting System – lighting system allows adjustment of light and removing noise. LEDs are used with white light with a wavelength of 450nm.  Matlab procedure for image processing and analysis is used. The algorithm is presented in Table 1. Table 1. Algorithm for image processing of eggs from Japanese quails Step Working stage Description Image is photographed and then filtered. Preliminary image processing and Image is converted from RGB to HSV color A collection of information about model. Separation of H, S and V color colors components. Segmentation of the image based Determination of the gradient of image by B on the color threshold segmentation. C Output histogram of the image Imhist function Number of pixels, mean, standard deviation, D Display information about the image minimum and maximum value of the color components in the histogram are displayed.

ARTTE Vol. 1, No. 3-4, 2013 ISSN 1314-8788 (print), ISSN 1314-8796 (online) 182

ARTTE 3.

Applied Researches in Technics, Technologies and Education

Journal of the Faculty of Technics and Technologies, Trakia University https://sites.google.com/a/trakia-uni.bg/artte/

RESULTS AND DISCUSSION

Figure 1 shows the histograms of R, G and B components of the RGB color model to the image for five types of Japanese quail eggs. As can be seen from the histograms presented RGB color model is not suitable for utilization definitions color decoration of the shell of the eggs because they are multimodal and require more complex methods for assessment.

R G B Figure 1. Histograms of R, G and B component (RGB color model) for the five types of Japanese quail eggs The histograms of the images of eggs for three components of the HSV color model are displayed in Figure 2. There is a significant difference in the histogram of the S component ie through this color parameter can be distinguished different types of eggs of Japanese quails by their decoration. Suitable for a particular task proven HSV model of the visual image and in particular its Scomponents (Saturation). In the MATLAB environment its value is amended in the range 0÷1.

H S V Figure 2. Histograms of the H, S and V components (HSV color model) for five types of eggs from Japanese quails

ARTTE Vol. 1, No. 3-4, 2013 ISSN 1314-8788 (print), ISSN 1314-8796 (online) 183

ARTTE

Applied Researches in Technics, Technologies and Education

Journal of the Faculty of Technics and Technologies, Trakia University https://sites.google.com/a/trakia-uni.bg/artte/

In this case the cumulative property for the S-component of the model is used. The property consists of an examination of the number of pixels with a certain value, the resulting number is charged to results already obtained for the previous value of pixels. Figure 3 shows the S-component histograms of two types of eggs. The p-level between these histograms is p=0.007 and the difference is statistically significant.

Mean=0.27; SD=±0.03

Mean=0.39; SD=±0.04

Figure 3. Histograms of S component (HSV color model) for two types of eggs 4.

CONCLUSION

In this report a method for evaluating the quality of the eggs of Japanese quails by computer vision is presented. An algorithm for image analysis of quail eggs shell color is proposed. The statisticall signifficant difference between S-components shows that HSV color model is more suitable for evaluation of the images relative to RGB. For the purpose of the study is developed algorithm and software program for image processing of eggs from Japanese quail in Matlab. In implementation of HACCP system for safety management and quality control is possible wider application of computer vision systems. HSV color model is a mean of contactless automatic control at critical control points in fault detection and determination of quality parameters of eggs. 5.

REFERENCES

[1] A simple guide to understanding and applying the Hazard Analysis Critical Control Point Concept (HACCP), ILSI Europe, Avenue E. Mounier 83, Box 6, B-1200 Brussels, Belgium. [2] Dudusola I. O. Comparative evaluation of internal and external qualities of eggs from quail and guinea fowl. International Research Journal of Plant Science, 1 (5), 2010, pp. 112-115. [3] Genchev А., G. Mihaylova, S. Ribarski, A. Pavlov, M. Kabakchiev. Meat quality and composition in Japanese quails. Trakia Journal of Sciences, Vol. 6, No. 4, 2008, pp.7282, ISSN 1312-1723 (print), ISSN 1313-3551 (online).

ARTTE Vol. 1, No. 3-4, 2013 ISSN 1314-8788 (print), ISSN 1314-8796 (online) 184

ARTTE

Applied Researches in Technics, Technologies and Education

Journal of the Faculty of Technics and Technologies, Trakia University https://sites.google.com/a/trakia-uni.bg/artte/

[4] Lazarov I., Z. Zlatev, M. Mladenov. Analysis of the applicability of computer vision systems in the assessment of health risks of food quality. Conference ECO Varna "Transport, Ecology, Sustainable Development", Varna, 2013, ISBN 954-20-00030. [5] Mladenov M. I. Intelligent sensors and systems. ISBN 978-954-712-493-6, Ruse, 2010 (the original is in Bulgarian). [6] Punia B. K., B. G. Ramesh, M. P. Gnana, A. R. Rajasechar. A study of egg quality traits in Japanese quails. J. Veterinary & Animal Sciences 4 (6), 2008, pp. 227-231. [7] Regulation (EC) № 1028/2006 of 19 June 2006 on marketing standards for eggs (the original is in Bulgarian). [8] Regulation № 1 of 09.01.2008 on the requirements for eggs for consumption (the original is in Bulgarian). [9] Song K. T., S. H. Choi, H. R. Oh. A comparison of Egg Quality of Pheasant, Chukar, Quail and Guinea Fowl. Asian-Aus. J. Anim. Sci., 13 (7), 2000, pp. 986-990.

ARTTE Vol. 1, No. 3-4, 2013 ISSN 1314-8788 (print), ISSN 1314-8796 (online) 185