Detection Of Texture Characteristics Of Beef Base On Biospeckle Technique

As the development of the economy and the optimization of the consumption structure,consumption of beef got rapid growth in China.Therefore,beef quality has been put more attention on by meat industry,and it was usually evaluated by expert judgment and mechanical equipment.The expert judgment could ensure lower classification and non-destructive grading process,but it might induce inconsistent results by different persons.Meanwhile,the mechanical equipment could be an option to estimate meat quality;However,the problems exiting in these tests are that they are time consuming,costly and destructive.The dynamic laser speckle is a portable,non-invasive and low-cost technique,which has been widely used for agricultural products.In order to detect the quality of beef fastly,the condition of biospeckle had been optimized,the parameters of models had been chosen.And the conclusions of this paper were shown as follows:The single factor experimental design was first applied for determining preliminary suitable lighting conditions for the later Box-Behnken experimental design.Subsequently,a response surface regression function of three lighting factors was obtained to optimize the measurement conditions for beef dynamic biospeckle.Above results demonstrated that the speckle activity could reach 476.04,the R² value was0.992,Root Mean Square Error value was 8.14,Bias Factor?B_f?and Accuracy Factor?A_f?values were within the acceptable range of 1.0<B_f<A_f<1.05 via the optimal processing.Therefore,for the further beef biospeckle measurement,the lighting conditions were as follows:lighting intensity was 30 mW,laser wavelength was 635 nm and laser incident angle was 15?.Though the research of different texture measurement methods,the results showed that the hardness and chewiness of sensory evaluation and TPA had high positive correlation,the R² reached 0.98 and 0.90 respectively,and R² between W-B shear force and hardness of TPA reached 0.95,which proved the reliability of three texture characteristics measurements.In order to predict the texture characteristics viz biospeckle activity,the R² of hardness,chewiness and W-B shear force were 0.83,0.77and 0.69,respectively.The results of improvement for the loin model were as follows:as improved with S/B correction method,the R_MSE was 26.65,B_f and A_f values were 1.08and 1.15,respectively.Meanwhile,the effect of modified with K-S adding method of typical samples was better,and when the adding number of samples were 12,the R_MSE was 13.21,B_f and A_f values were 1.07 and 1.02,respectively.In conclusion,K-S typical samples adding method could reduce the differences between the different parts,improve the goodness-of-fit of predictive shin model,and produce better effect than S/B correction method.Three-dimensional biospeckle was first applied in prediction of beef texture to minimize time and reduce the limitations of its application.The images of samples were shoot from two different angles,preproccessed by GD?Generalised different?,and fused by gray template matching method,wavelet transform and contrast modulation fusion.The traditional camera calibration method was used,and the information of samples depth was extracted by similar triangle principle.The results showed that three-dimensional biospeckle could predict beef texture better.Compare the three method of images fusion,it turned out that the effect of wavelet transform was best,the correlation coefficient of prediction for hardness and chewiness reached 0.944 and0.929,respectively.The above results demonstrate that biospeckle could predict texture character of beef well after condition optimization;through the improvement of modeling,the texture ofdifferent parts of beef could be predicted,the resources needed for the reconstruction model could be saved greatly;compared with traditional biospeckle,the three-dimensional biospeckle could reduce the time consuming of image acquisition and adjust the depth information of the beef samples.