Font Size: a A A

Research And Implementation Of Pig Breeding Optimum Selection Method Based On CT Image

Posted on:2020-06-27Degree:MasterType:Thesis
Country:ChinaCandidate:J WangFull Text:PDF
GTID:2393330578470491Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
The research on breeding technology of pigs is in an important position all over the world.At present,the breeding technology of pigs is dominated by the cultivation of specialized strains and the cultivation of supporting systems.More scholars have tried to introduce other technologies into breeding in order to develop new breeding programs,and thus the application of medical images for breeding technology has emerged.With the development of medical image field,medical image combined with computer technology has become an important research direction.Medical image segmentation,as the basis of medical image processing and analysis,has undergone manual segmentation,semi-automatic segmentation and full-automatic segmentation.It has been applied to bone segmentation,lung extraction,tumor removal and so on in human body.The segmentation method of human structure used in medical image was also applicable to animal body,but no segmentation algorithm could be applied to different structural objects.To solve this problem,this paper was based on the CT image of the preferred method of breeding pigs,processed a segmentation method that accurately divides bone,fat,and muscle tissue and determined its mass fraction in the whole.The method minimized human intervention,quickly and accurately segmenting parts of bone,fat,muscle and determining their proportion.This paper took the animal body as research object and the acquired CT images were analyzed.Firstly,CT images were preprocessed,the Gaussian filter was used to smooth the image and eliminate the noise,and then the CT bed which affected the bone segmentation in the CT image was removed by using the idea of region growing.After removal,two segmentation methods were used to extract the bone parts,which were the fast level set algorithm based on the improved level set and the threshold segmentation method based on the bone CT value.Considering the presence of visceral parts had a certain influence on fat and muscle segmentation,visceral removal was carried out by using the proportional idea of organ distance in the body,combined with threshold algorithm,morphological operation,hole filling and other algorithms.Then the fat and muscle parts were segmented by double threshold method and logical operation.Finally,the quality and the proportion of each part in the whole body were calculated according to the volume and density of the pixels in the body.The algorithm was applied to the experiment and the possible problems in the algorithm were explored by comparing with the comparative data provided by the US company,so as to further improved the algorithm as the focus of future research.
Keywords/Search Tags:medical image segmentation, Breeding, fast level set, visceral removal
PDF Full Text Request
Related items