Freshness detection of Veal using image processing and Response Surface Method

مجله پژوهشهای علوم و صنایع غذایی ایران

Article 26, Volume 13, Issue 2 - Serial Number 44, May and June 2017, Pages 251-261   

Document Type: Research Article

DOI: 10.22067/ifstrj.v1395i0.50161

Authors

Hossein Javadi kia^{* 1}; Mahdi Ghasemi-Varnamkhasti²; Sajad Sabzi¹

¹Department of Mechanical Engineering of Biosystems, Razi University, Kermanshah, Iran.

²Department of Mechanical Engineering of Biosystems, Shahrekord University, Shahrekord, Iran.

Abstract

Introduction: Nowadays, with the development of imaging systems and image processing algorithms, a new branch of agriculture and food industry quality control has emerged. Meat and related products have high commercial value and they are one of the most important items of household food basket (Jackman et al., 2011). The apparent color of the meat is one of the most important ranking factors which determines the quality and marketing value (Ramirez and Cava, 2007; Shiranita et al., 2000). There is a relationship between color, appearance etc. and the shelf-life of meat, since the passing time causes the color to be darkened in meat due to chemical reactions and shrinkage occures Therefore, determining the storage time is important in terms of quality and marketing value (Jackman et al., 2011; Tan, 2004). In recent years, virtual image on a computer as a helpful suggestion for meat grading has been emerged. Various studies have been conducted in the field and results in a number of applications suggests that the color image processing method for assessing the quality of meat is important (Girolami et al., 2013; Mancini and Hunt, 2005; Lu et al., 2000). Due to the importance of detection of freshness veal in order to preserving the quality and post ponding the meat spoilage and disease accordingly, designing a device to detect the storage time of slaughter and in other words the freshness of veal using image processing and response surface method was studied. For this purpose, two common environment and standard maintenance of fresh meat: first in the refrigerator with an average temperature of 3°C and second in cool place with a temperature of 8°C were considered and then the effects of storage time on the meat quality was observed using a digital camera Some common models were developed for image processing and the response surface method was applied.

Materials and methods: First, some meat from three sections of veal meat: hands, feet and neck, were prepared from Kermanshah slaughterhouse and the slaughtered time was recorded as an initial time. From each of the six states in total, 18 samples were taken appropriate to the thickness of one centimeter. Samples were randomly selected for inclusion in the standard conditions (ISIRI 692).
Image processing:
More than 600 images were acquired at various storage times and they were then evaluated to find the appropriate separation methods for meat from image background. The best way to separating the meat image from the background in the image was using the RGB color and the B space values with 150 value as the threshold. In other words, the exact coordinates of meat pixels were obtained. Then background isolated by edge detection with Cany filter with coefficient of 0.7. Finally meat image was isolated from background. Then various parameters of meat image were extracted. The number of parameters were more than 50 parameters. Then sensitivity analysis were selected as three parameters: Contrast, Roughness, and Texture that had more influence on time change from the moment of slaughter and were selected as appropriate inputs of models.
Modeling by Response Surface Method:
In this method, selected parameters were used as inputs and the time of slaughter in minutes, was used as output of the model. Because of the more difference of the values of various parameters from each other, all data were normalized. Generally due to the three organs of veal and two different environments to maintain, six models in the Software Design Expert 7.0 were designed and optimized using response surface methods. In the next step, data samples at ambient temperature as well as refrigerated samples were modeled.

Results and discussion: The results of the models by the response surface methods were good and acceptable. In the final step the general models were good; these models were about all of data in environment and refrigerator.

Conclusion: In this study, considering the importance of using fresh meat calves by people as well as processing plants, some algorithms were designed and developed to estimate the pasted time of the calf slaughtered. For this purpose Samples were prepared from three parts of slaughtered calves: ham, shoulder and neck. The samples were stored in the environment and common standards place the first in refrigerator with a temperature of 3 ° C and another in cool environment with an average temperature of 8 ° C. Then some images were taken from samples at specified times. Then some parameters were extracted from images produced by the image processing in MATLAB. Then by response surface method was designed and optimized. Suitable models and finaly suggested device has ability to estimate the time of slaughter by taking image.

Keywords

Veal; Image processing; Response surface; freshness meat; time of slaughter

References

Adelkhani, A., Beheshti, B.,Minaei, S., Javadikia, P. & Ghasemi-Varnamkhasti, M., 2013, Taste characterization of orange using image processing combined with ANFIS. Measurement, 46(9), 3573-3580.

Chmiel, M., Słowiński, M. & Dasiewicz, K., 2011-a, Lightness of the color measured by computer image analysis as a factor for assessing the quality of pork meat. Meat Science, 88, 566-70.

Chmiel, M., Słowiński, M. & Dasiewicz, K., 2011-b, Application of computer vision systems for estimation of fat content in poultry meat. Food Control, 22,1424-1427.

Girolami, A., Napolitano, F., Faraone, D. & Braghieri, A., 2013, Measurement of meat color using a computer vision system. Meat Science, 93, 111-118.

ISIRI 692, Meat, chicken, egg and fish-storage in cold stores - Code of practice, Institute of Standards and Industrial Research of Iran, 1st. revision.

Jackman, P., Sun, D.-W. & Allen, P., 2011, Recent advances in the use of computer vision technology in the quality assessment of fresh meats. Trends in Food Science & Technology, 22, 97-185.

Lu, J., Tan, J., Shatadal, P. & Gerrard, D., 2000, Evaluation of pork color by using computer vision. Meat Science, 56, 57-60.

Mancini, R. & Hunt, M., 2005, Current research in meat color. Meat Science, 71, 21-100.

Monin, G., 1998, Recent methods for predicting quality of whole meat. Meat Science, 49, 43-231.

Ramirez, R. & Cava, R., 2007, The crossbreeding of different Duroc lines with the Iberian pig affects colour and oxidative stability of meat during storage. Meat science, 77, 47-339.

Shiranita, K., Hayashi, K., Otsubo, A., Miyajima, T. & Takiyama, R., 2000, Grading meat quality by image processing. Pattern Recognition, 33, 97-104.

Tan, J., 2004, Meat quality evaluation by computer vision. Journal of Food Engineering, 61, 27-35.

Valous, N.A., Mendoza, F., Sun, D.-W. & Allen, P., 2009, Colour calibration of a laboratory computer vision system for quality evaluation of pre-sliced hams. Meat Science, 81, 41-132.

Wu, D. & Sun, D.-W., 2013, Colour measurements by computer vision for food quality control – A review. Trends Food Sci Technol, 29, 5-20.

Zheng, C., Sun, D.-W. & Zheng, L., 2006, Correlating colour to moisture content of large cooked beef joints by computer vision. J Food Eng, 77, 63-858.