Unlock instant, AI-driven research and patent intelligence for your innovation.

Image interlayer boundary determination method and device based on neural network

A technology of neural network and determination method, which is applied in the field of image segmentation, can solve problems such as misjudgment of layered boundaries and inability to determine the confidence level of layered boundaries, so as to improve accuracy, reduce complexity and difficulty of overall debugging, and eliminate inaccuracies. The effect of a point with a high degree of certainty

Pending Publication Date: 2021-10-22
唯智医疗科技佛山有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it has been found in practice that this method cannot determine the confidence level of the identified layer boundaries, and there is a risk of misjudgment of the layer boundaries

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 interlayer boundary determination method and device based on neural network
  • Image interlayer boundary determination method and device based on neural network
  • Image interlayer boundary determination method and device based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0079] see figure 1 , figure 1 It is a schematic flowchart of a neural network-based method for determining boundaries between image layers disclosed in an embodiment of the present invention. Wherein, the method can be applied to an image layer determination device, and the image layer determination device can be an independent device, or can be integrated in a picture or video processing device, which is not limited in this embodiment of the present invention. Such as figure 1 As shown, the method for determining boundaries between image layers based on a neural network may include the following operations:

[0080] 100. Input the image to be detected into the first convolutional neural network model to obtain a predicted boundary result, the predicted boundary result including N layered boundaries and N boundary channels corresponding to the layered boundaries.

[0081] In the embodiment of the present invention, the predicted boundary result of the image to be detected ...

Embodiment 2

[0109] see figure 2 , figure 2 It is a schematic flowchart of another neural network-based method for determining boundaries between image layers disclosed in an embodiment of the present invention. Wherein, the method can be applied to an image layer determination device, and the image layer determination device can be an independent device, or can be integrated in a picture or video processing device, which is not limited in this embodiment of the present invention. Such as figure 2 As shown, the method for determining boundaries between image layers based on a neural network may include the following operations:

[0110] 201. Input the image to be detected to the first convolutional neural network model to obtain a predicted boundary result, the predicted boundary result including N layered boundaries and N boundary channels corresponding to the layered boundaries.

[0111] 202. For any boundary position in any boundary channel, determine a target pixel point set corr...

Embodiment 3

[0138] see image 3 , image 3It is a structural schematic diagram of a device for determining boundaries between image layers based on a neural network disclosed in an embodiment of the present invention. It should be noted that the device for judging movement and stillness refers to the steps in the neural network-based method for determining the boundary between image layers described in Embodiment 1 and Embodiment 2, and the detailed description will not be repeated in this embodiment. For example, image 3 As shown, the device for determining the boundary between image layers based on the neural network may include:

[0139] The first processing module 301 is configured to input the image to be detected into the first convolutional neural network model to obtain a predicted boundary result, the predicted boundary result including N layered boundaries of the image to be detected, corresponding to the N layered boundaries The N boundary channels and the probability value ...

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 interlayer boundary determination method and device based on a neural network; the method comprises the steps: inputting a to-be-detected image into a first convolutional neural network model, and obtaining a prediction boundary result which comprises N layered boundaries and N boundary channels corresponding to the layered boundaries; for a boundary position in any boundary channel, determining a target pixel point set corresponding to the boundary position; calculating the distribution concentration degree of the probability values of all the pixel points in the target pixel point set belonging to the boundary channel, and obtaining the distribution concentration degree corresponding to the boundary position; and determining the certainty degree information of the boundary position according to the distribution concentration degree corresponding to the boundary position. The method can accurately segment the layering of the image based on the neural network and obtain the layering boundary information, thereby determining the accuracy information of the recognized boundary information, and facilitating the improvement of the recognition accuracy of the layering boundary of the image.

Description

technical field [0001] The invention relates to the technical field of image segmentation, in particular to a neural network-based method and device for determining boundaries between image layers. Background technique [0002] With the rapid development of computer and medical technology, OCT (Optical Coherence Tomography, optical coherence tomography) technology has been widely used in the diagnosis equipment of fundus diseases, which is of great significance to the detection and treatment of ophthalmic diseases. OCT belongs to a high-sensitivity, high-resolution, high-speed, non-invasive tomographic imaging method, which uses the coherence of light to scan and image the fundus. Each scan is called an A-scan, and adjacent consecutive multiple scans The combination is called a B-scan, and the B-scan is the commonly seen OCT cross-sectional view, which is the most important imaging method of OCT in medical diagnosis. [0003] In the practical application of analyzing and pr...

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): G06T7/12G06N3/08G06N3/04
CPCG06T7/12G06N3/084G06N3/045
Inventor 区初斌叶重荣彭勇
Owner 唯智医疗科技佛山有限公司