Method and device for segmenting object from image

An image and object technology, applied in the field of image recognition, can solve problems such as difficult to express, and achieve the effect of fast and accurate segmentation

Inactive Publication Date: 2015-03-25
BEIJING SAMSUNG TELECOM R&D CENT +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Existing technologies often can only extract features within the basic unit, but it is difficult to express the overall shape and structural features of the target

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for segmenting object from image
  • Method and device for segmenting object from image
  • Method and device for segmenting object from image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] Embodiments according to the present invention will be described below with reference to the drawings.

[0025] figure 1 A flowchart showing a method of segmenting an object from an image according to an embodiment of the present invention.

[0026] Such as figure 1 As shown, in step 101, an object detector is used to detect a predetermined number of target object regions from an image.

[0027] Object regions may be detected using an object detector pre-trained to detect objects to be detected. For example, assuming that a certain tissue of the human body is an object to be detected, an object detector for detecting the tissue may be pre-trained. The object detector can be trained using various existing object detector training techniques, for example, the object detector can be implemented by using a Boosting classifier, a variable part (DPM) model, a support vector machine (SVM) classifier, and the like.

[0028] The images here may be various images such as ultr...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

Provided are a method and a device for segmenting an object from an image. The method comprises the steps of using an object detector to detect a predetermined number of target object regions from an image, acquiring the features of each target object region, segmenting the image into a plurality of grids, acquiring the features of each grid and the connection features between the grids, and determining an object in the image according to the potential calculated based on the features of the target object regions, the features of the grids and the connection features. By adopting the method and the device for segmenting an object from an image provided by the invention, image information of various levels can be effectively utilized to quickly and accurately segment a target object.

Description

technical field [0001] The invention relates to the field of image recognition. More specifically, it relates to a method and apparatus for segmenting objects from images. Background technique [0002] Image segmentation techniques are used to extract objects from images. For example, image segmentation technology can be used to extract the contours of tissues or tumors from ultrasound images or medical images of other modalities. [0003] Currently, there are mainly two types of image segmentation technologies, one is top-down (Top-down) image segmentation technology, and the other is bottom-up (Bottom-up) image segmentation technology. [0004] Top-down image segmentation technology mainly uses high-level (High-level) image information (for example, the structure of the target object, contour features, etc.) to extract the target object. Usually, the approximate location of the target is determined first (for example, the bounding box of the target object is detected by...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00
CPCG06T7/11G06T2207/20112
Inventor 郝志会王强王晓涛郭萍金培亭
Owner BEIJING SAMSUNG TELECOM R&D CENT
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap