首页> 外国专利> ULTRASOUND SYSTEM FOR FORMING 3D FETUS ULTRASOUND IMAGE BASED ON FETUS SURFACE IMAGE EXTRACTED BY SVM-BASED TEXTURE CLASSIFICATION AND METHOD FOR THE SAME

ULTRASOUND SYSTEM FOR FORMING 3D FETUS ULTRASOUND IMAGE BASED ON FETUS SURFACE IMAGE EXTRACTED BY SVM-BASED TEXTURE CLASSIFICATION AND METHOD FOR THE SAME

机译:基于基于支持向量机的纹理分类提取的胎面图像的超声三维超声图像形成系统及其方法

摘要

An ultrasound system for forming a 3D fetus ultrasound image based on a fetus surface image extracted by SVM(Support Vector Machines)-based texture classification and a method for forming the same are provided to automatically extract a fetus surface from the 3D fetus ultrasound image without interference of a user. An ultrasound system for forming a 3D fetus ultrasound image based on a fetus surface image extracted by SVM-based texture classification includes an ultrasound diagnosing unit(10) providing a 3D ultrasound image of an object and a periphery of the object, wherein the 3D ultrasound image is composed of a plurality of frames. A pre-processing unit(20) selects a predetermined number of key frames among the plurality of frames. A dividing unit(30) divides an object area and a non-object area in the key frame. A texture-based area classification and combining unit(40) classifies and combines the divided object and non-object areas into the object area and the non-object area based on texture. A surface extraction unit(50) removes the non-object area classified based on texture from each key frame, combines the adjacent object areas in the key frame to form an object image, and then extracts a contour of a front surface of the object body from the object image of each key frame. A 3D rendering unit(60) determines a 3D surface by matching and connecting the front surface of the object body extracted from each key frame, forms a masked volume composed of the determined 3D object body surface, and then performs rendering of the masked volume to form the 3D ultrasound image.
机译:提供了一种基于通过基于SVM(支持向量机)的纹理分类所提取的胎儿表面图像来形成3D胎儿超声图像的超声系统及其形成方法,以自动地从3D胎儿超声图像中提取胎儿表面。用户的干扰。用于基于通过基于SVM的纹理分类提取的胎儿表面图像来形成3D胎儿超声图像的超声系统包括超声诊断单元(10),超声诊断单元提供对象的3D超声图像和对象的外围,其中3D超声图像由多个帧组成。预处理单元(20)在多个帧中选择预定数量的关键帧。划分单元(30)在关键帧中划分对象区域和非对象区域。基于纹理的区域分类和合并单元(40)基于纹理将划分的对象和非对象区域分类和合并为对象区域和非对象区域。表面提取单元(50)从每个关键帧中去除基于纹理分类的非对象区域,组合关键帧中的相邻对象区域以形成对象图像,然后提取对象主体的前表面的轮廓。从每个关键帧的对象图像。 3D渲染单元(60)通过匹配并连接从每个关键帧提取的物体的前表面来确定3D表面,形成由所确定的3D物体表面构成的掩蔽体,然后执行该掩蔽体的渲染。形成3D超声图像。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号