Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Image collaborative cutout method based on confidence level

A confidence and image technology, which is applied in the field of image collaborative matting based on confidence, can solve the problems that cannot be established, the real scene image does not completely match, and can not get better results, etc.

Inactive Publication Date: 2014-07-23
NANJING UNIV
View PDF4 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In the progress of matting technology in recent years, researchers have tried a variety of techniques, but in the matting of real natural scene images, complete effects are still not obtained. The main reason is that real scene images do not fully conform to the three main categories. Assumptions and Premises of Cutout Technology
In the sampling-based method, if the color distribution of the foreground F and the background B have a large overlap, the sampling-based method cannot sample a suitable foreground and background pixel pair for the pixel to be solved, resulting in poor results; In the propagation method, the assumption of the Local Color Line Model (Local Color Line Model) cannot be established in the high-gradient edge texture under the real scene image, so that better results cannot be obtained.
In response to such problems, the existing method is generally to further provide a more accurate mask, but this undoubtedly increases the amount of manual work

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
  • Image collaborative cutout method based on confidence level
  • Image collaborative cutout method based on confidence level
  • Image collaborative cutout method based on confidence level

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0118]Since the present invention processes images, it is unavoidable to use a grayscale image in the drawings showing the processing process and the effect of the embodiment.

[0119] figure 2 Two source images to process are given. It can be seen that the two images have a similar foreground and a large change in the background.

[0120] image 3 The masks of the two images obtained after steps 1-3 are given. Among them, area 1 is the foreground, area 2 is the area to be solved, and area 3 is the background.

[0121] Figure 4 For the result after using the matting algorithm based on global sampling, it can be seen that some of the matting results where the foreground and background are mixed are not very good.

[0122] Figure 5 It is a confidence map obtained after feature extraction and regression analysis. Different colors reflect different confidence levels. The lighter the color, the higher the reliability. Correspondingly, it tends not to change during the coll...

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

The invention discloses an image collaborative cutout method based on confidence level. The method aims to conduct collaborative cutout on images with a foreground of slight deformation and a background of large difference. According to the method, front background segmentation is conducted on multiple images by the adoption of a collaborative segmentation algorithm, each image is marked as a mask of a foreground, a background and a region to be solved through morphological operation, cutout is conducted on each source image by the utilization of the source image and the mask by the adoption of an existing single common cutout method, confidence measurement is conducted on the cutout result, pixel points in the region to be solved in all the images are matched, a global optimization function is defined based on the matching so that the cutout effect on all the images is improved in coordination, the method aims to conduct cutout result spread from a high confidence region to a low confidence region matched with the high confidence region, and thus more accurate cutout result is obtained in the corresponding low confidence region. By means of the method, multiple images are input, and the cutout result of the images is output.

Description

technical field [0001] The invention relates to an image processing method of computer vision, in particular to a confidence-based collaborative image-cutting method which has strong map-cutting ability and is relatively automatic, thereby saving a lot of manual interaction. Background technique [0002] The matting technology aims to accurately separate the foreground and background of a natural scene image. Given a natural scene image, according to the linear composite formula: [0003] I=αF+(1-α)B [0004] The matting technology represents the pixel value I in a natural image as a linear composite of the foreground F and background B of the image and the foreground transparency α (also called alpha matte), where the value of α is between 0 and 1. [0005] The matting technology has been widely used and achieved great success in the industry, especially in the image processing and film industries. Therefore, matting technology has been an important research content of com...

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
IPC IPC(8): G06T7/00
Inventor 汪粼波夏天辰郭延文
Owner NANJING UNIV
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
Eureka Blog
Learn More
PatSnap group products