Font Size: a A A

Research On Cattle Follicle Monitoring System Based On 3-D Ultrasound Image

Posted on:2017-10-14Degree:MasterType:Thesis
Country:ChinaCandidate:H W JiangFull Text:PDF
GTID:2323330485950473Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Ultrasound medical images have been widely used in many field with its fast,inexpensive real-time,safety and low cost advantages.Among them,ultrasound imaging technology as a new technology used in cattle breeding and embryo production is an effective means to improve cattle breeding,based on monitoring and analysis of dynamic three-dimensional ultrasound follicles to improve cattle pregnancy has great theoretical and practical significance.Currently,based on three-dimensional ultrasound image processing technology cattle follicle contains research on three-dimensional ultrasound images speckle reduction,follicle detection and segmentation,follicle volume measurement.Compared with the traditional two-dimensional ultrasound,three-dimensional ultrasound images have intuitive display,accurate positioning,efficiency,structural parameters measurable and many other advantages,using three-dimensional ultrasound image processing technology can greatly improve the process of reproduction in cattle follicular dynamic changes the accuracy of detection,three-dimensional ultrasound image processing technology plays a very important role.This article focused on follicle speckle reduction,edge detection,3D reconstruction technique research,the speckle reduction algorithm and its implementation on CUDA acceleration based on anisotropic diffusion,edge detection algorithm based on extended class Harr features and gradient histogram with AdaBoost,as well as reconstruction algorithm based on point cloud of the three-dimensional image of follicles,finally introduced the Cattle Follicle Monitoring System based MFC and OpenGL technology.
Keywords/Search Tags:Three-dimensional ultrasound image, Speckle reduction, Anisotropic diffusion, Edge detection, Machine learning, Three-dimensional reconstruction
PDF Full Text Request
Related items