The Application Of Computer Vision Technology In Evaluation Of Meat Quality

As a non-destructive measurement method, computer vision technology has been widely applied in food detection field. However, the research on evaluation of carcass and chilled meat quality has not been conducted yet in China. This paper discussed the application of computer vision technology in evaluation of carcass classification and chilled meat quality, which includes marbling distribution and meat color.Based on above goals, this paper finished the following works:1.The carcass image analysis software is programmed using Visual C++, which can perform image acquisition, system calibration , morphological features extraction and length scale conversion.2.The carcass image and 6~7th cross section image have been analyzed in Visual C++ and MATLAB environment. The information extracted and measured were three-point fat thickness, rib eye area and carcass length. The carcass cutability predicted equation using SAS (8.1) STEP WISE method isy=32.6354+0.1718× eye areaa +0.0654× carcass weight -1.4321× 2-point fat thicknessa -2.5798× 6-7th fat thicknessa (R2=0.75, p<0.01).3.Fat area ratio (FAR%) from image analysis and fat content(fat%) from chemical analysis were compared and the correlation analysis was made. The correlation between FAR% and fat% is very significantly, r=0.89 (p<0.01). The equation for the regression of the fat content (y) on FAR% (x) calculated for all of data is y=0.1845x-0.1397 with R2=0.78. The experiments show that computer vision technology can make an accurate evaluation about marbling.4.Color image feature were extracted from rib eye image by MATLAB software, and the color feature is compared with the sensory scores. Feature used in this study included mean and standard deviation of R,G, B, H, I and S bands of the rib eye area. The regression equation for predicted meat color isy=6.1375-8.3478× Hu+7.0459× (R-G) / (R+G) +0.1874×(2G-R-B) /2 (R2=0.55, p<0.01) .The objective of this research was to determine the potential of computer vision technology for evaluating carcass and chilled meat quality. Results of this research showed that computer system is an effective tool for evaluating meat quality.