Prediction of beef color standard number from digital image obtained by using photographing equipment for the cross section of carcass
2001
Kuchida, K. (Obihiro Univ. of Agriculture and Veterinary Medicine, Hokkaido (Japan)) | Hasegawa, M. | Suzuki, M. | Miyoshi, S.
The photographing equipment integrating the digital camera and the illumination device for the cross section of carcass was able to take a clear image. The first aim of this study was to improve the photographing equipment in order to adapt for more carcasses. The second aim was to predict the Beef Color Standard (BCS) number by using the obtained image from the equipment. Height of the photographing equipment was 12cm lower (H = 33cm) than the previous model with the adaptation of a ring illumination method instead of a coaxial epi-illumination method. The photographing equipment has 1,002 white LEDs (light emitting diodes) internally, which can irradiate a steady light to the cross-section of carcass uniformly. The correlation coefficients of R, 0, B components and luminance from image with BCS number of the standard model of BCS were -0.98, -0.91, -0.82, and -0.99, respectively. These results indicated that the relationship between luminance and BCS number of standard model was highly linear. Images around the rib eye at 6th-7th rib cross-section were collected (n = 1,208) at a meat processing plant in Hokkaido. BCS number assigned by grader was predicted by multiple regression analysis using image analysis traits. Ratios of the difference between BCS number assigned by grader and BCS number predicted by image analysis being =+- 0 and within =+- 1 were 76.8% and 100.0%, respectively.
显示更多 [+] 显示较少 [-]