Image segmentation method based on pseudo-color coding and DISCOV coding

A technology of pseudo-color coding and image segmentation, which is applied in the field of image segmentation based on pseudo-color coding and DISCOV coding, can solve the problems of slow calculation speed and low accuracy rate, achieve good segmentation effect, improve accuracy rate and calculation speed, and improve The effect of information

Inactive Publication Date: 2017-11-24
HARBIN INST OF TECH
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problems of low image extraction accuracy and slow calculation speed in the prior art under the condition of noise, and propose an image segmentation algorithm based on pseudo-color coding and DISCOV coding

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 segmentation method based on pseudo-color coding and DISCOV coding
  • Image segmentation method based on pseudo-color coding and DISCOV coding
  • Image segmentation method based on pseudo-color coding and DISCOV coding

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0033] Specific implementation mode one: combine image 3 Describe this embodiment, the specific process of the image segmentation method based on pseudo-color coding and DISCOV coding of this embodiment is:

[0034] Step 1. The simulation environment used in the present invention is the multi-angle aircraft data provided by the website https: / / github.com / kishankondaveeti / Synthetic_ISAR_images_of_aircrafts open source. The specific data used are 'a1_img026.mat', 'a5_img019.mat', 'a6_img087.mat'. Grayscale the data as input data; such as Figure 1a ;

[0035] Obtain image data, perform grayscale processing on the image data, obtain a grayscale processed image, add noise processing to the grayscale processed image, and obtain a noise-added image; as Figure 1b ;

[0036] The added noise satisfies the following conditions: the ratio of the signal power of the image to the power of the noise is 4;

[0037] Step 2, removing outliers from the noise-added image obtained in step ...

specific Embodiment approach 2

[0045] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that in the step 3, pseudo-color coding is performed on the image after removing abnormal values ​​to obtain a pseudo-color coded image; the specific process is:

[0046] Pseudo-color is a technique that replaces the gray value of pixels with color. It maps a monochrome image into a color image by matching each gray level to a point in the color space. Its mapping relationship can be expressed by the following formula:

[0047] The pseudo-color encoding based on the transformation of the pixel itself maps the grayscale image of the image after removing outliers to an RGB image, and the mapping relationship is:

[0048]

[0049] In the formula, f(x, y) represents the pixel gray value of point (x, y) in the image; T R ,T G ,T BIt is the mapping function between red, green, blue and gray respectively, which reflects the mapping relationship of pixels from gray to color. There ...

specific Embodiment approach 3

[0055] Embodiment 3: The difference between this embodiment and Embodiment 1 or 2 is that in the step 4, the pseudo-color-coded images R, G, and B are sent to DISCOV, and the pseudo-color-coded images are coded by single opposite elements. RG encoding; the specific process is:

[0056] The coding formula of a single opposite component is:

[0057] After encoding the grayscale image in pseudo-color, we have obtained rich color information that conforms to the human visual model. The method proposed in this paper can fully utilize and encode the color information to achieve further image understanding and cognition.

[0058] The DISCOV dimensionless shunt model converts RGB image elements into a model consisting of retinal, single-opposite element, and double-opposite element feature cascades, transforming the input into a computer-cognizable system.

[0059] Users can specifically establish a dimensionless shunt model based on the multi-channel information around the target an...

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 relates to an image segmentation method based on pseudo-color coding and DISCOV coding, and relates to an image segmentation method. The objective of the invention is to solve problems of low image extracting accuracy and low calculating speed under the noise condition in the prior art. The process comprises the steps: 1, obtaining image data, carrying out the gray scale processing of image data, obtaining an image after gray scale processing, carrying out the noise adding processing of the image after gray scale processing, and obtaining an image after noise adding; 2, eliminating abnormal values of the obtained image according to the 3sigma rule after noise adding, and obtaining an image after the abnormal values are eliminated; 3, carrying out the pseudo-color coding of the image after the abnormal values are eliminated, and obtaining a pseudo-color coding image; 4, inputting the R, G and B values of the pseudo-color coding image into DISCOV, and employing the RG of a single opposite element code for the coding of the pseudo-color coding image; 5, segmenting an RG coding coefficient into 1 or 1 through the otsu method. The method is used in the field of image segmentation.

Description

technical field [0001] The invention relates to an image segmentation method. Background technique [0002] Traditional target area extraction methods such as fuzzy C classification can adjust the threshold according to the cohesion inside the target and the Euclidean distance between the target and the background to achieve the extraction of the target area. In recent years, a series of image segmentation methods based on retinal models have been continuously proposed, hoping to deal with complex and changeable background environments by simulating the processing of the human retina. Image segmentation methods based on retinal models include pulse-coupled neural network (PCNN), intersecting visual cortex (ICM), and improved intersecting visual cortex (SICM). PCNN is a three-layer oscillating network with two convolutions and five equations. Feedback neurons, connection neurons, internal drives, and dynamic thresholds are constantly adjusted to make it constantly approach t...

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): G06T7/10G06T9/00
CPCG06T7/10G06T9/00G06T2207/10024
Inventor 李杨贺梦珂张宁
Owner HARBIN INST OF TECH
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