Estimation of the chicken meat freshness using of color image processing and response surface methods

The importance of chicken meat as a safe and nutrient food product is considerable for people all over the world. So checking of its quality has the great value. Therefore, the main objective of the current study is the diagnosis of chicken meat freshness using the estimation of elapsed time from slaughter helping of image processing and response surface methods. In order to achieve this goal, chicken thighs were selected as the case study and they were stored in the fridge temperature and desired images were prepared at the specified times. After that statistical features of texture images were extracted of different color channels, by application of sensitivity analysis method, effective features were selected in elapsed time from slaughter. At the end response surface method was applied to design and optimize the regression models in order to estimate the elapsed time from slaughter. The applied statistical indicators for validation of optimized regression models include R-Squared ، Adj R-Squared، Pred R-Squared، RMSE and Press RMSE. The value of these indicators for the with skin part of chicken meat (optimized) were obtained 0.901, 0.899, 0.898, 27.31 and 27.44 and for the skinless part of chicken meat (optimized) were 0.866, 0.865, 0.864, 29.66 and 29.7. The acceptable obtained results indicate that image processing and response surface methods have the ability to diagnosis the elapsed time from slaughter as well.