Key Technologies Of Image Processing Of Bone Fragments In Meat Products

The detection of bone fragments in deboned meat products concerns the food industry. As an efficient and reliable non-destruction technology, X-ray has been applied in detecting bone fragments. The difference in X-ray absorption due to uneven distribution of meat thickness produces false patterns in X-ray images, interfering with the detection of bone fragments. Measuring the thickness information of meat products with laser triangulation to compensate the false patterns in X-ray images is adopted.The thickness-compensation of the meat and the meat products is studied in this paper. The constitutions of X-ray imaging system and laser thickness-measurement are introduced, and the principles of them are analyzed. The factors of affecting image quality in image-forming system and laser thickness-measurement are studied. The optimal parameters of X-ray and the angle of laser thickness-measurement are chosen. This paper focused on the pretreatment of X-ray images and laser thickness-curve images, including image denoising, histogram equalization, etc. The laser thickness-curves are refined and converted to laser thickness-image, and the experimental results are analyzed.The fusion methods of X-ray images and laser thickness-image are studied after laser thickness-image registration and gray-scale calibration. According to image fusion methods which are based on wavelet transform, the corresponding improved measures are proposed in this paper. The experiment results indicate that the performances of the image fusion methods are better than the performances before the image fusion method were improved.This paper gives the simulation test on the above image processing in matlab7.0 environment. The experiment results indicate that through the research of this topic can be effectively altered the deviation of X-ray gray image which occur due to the uneven thickness of meat, and false alarm and misdetection could be decreased effectively, consequently, making the detection of bone fragments more scientific, standardized and intelligent.