Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Data preprocessing method and system based on image feature refinement

A data preprocessing and image feature technology, applied in the fields of instruments, character and pattern recognition, computer parts, etc., can solve the problem of insufficient training data acquisition, avoid the collection of invalid data, ensure accuracy, and improve the effect of deep learning

Inactive Publication Date: 2019-09-27
GUANGDONG KINGPOINT DATA SCI & TECH CO LTD
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a data preprocessing method and system based on image feature refinement, which can avoid the problem of insufficient training data in existing image processing methods

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
  • Data preprocessing method and system based on image feature refinement
  • Data preprocessing method and system based on image feature refinement
  • Data preprocessing method and system based on image feature refinement

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] like figure 1 Shown: A data preprocessing method based on image feature refinement, including the following steps:

[0046] S1: The step of calculating the effective area, performing mask processing on the original image to obtain the training image, and calculating the effective area of ​​the training image. The effective area includes the positive sample area and the negative sample area. The value of the effective area is calculated through opencv mask processing. The positive sample in the positive sample area is manually marked by a professional doctor, and the negative sample in the negative sample area is the area other than the positive sample.

[0047] S2: training data collection step, constructing a sliding window, moving the position of the sliding window on the training picture, and collecting training data when the digital center of the sliding window is located at the boundary or inside of the effective area.

[0048] S3: In the deep learning step, the c...

Embodiment 2

[0063] The embodiment also discloses a data preprocessing system based on image feature refinement.

[0064] like Figure 7 Shown: a data preprocessing system based on refinement of image features, including a collection terminal and a server. The collection terminal is wirelessly connected to the server through an existing wireless communication module. In this embodiment, the wireless communication module uses VG's VG30S4T-X1 type wireless communication module. The function of the acquisition terminal can be realized through the existing medical microscope, and the server can be a Lenovo 2U rack server.

[0065] specific,

[0066] Collection terminals, including:

[0067] The image acquisition module is used to acquire medical original pictures. Existing medical microscopes are used to collect the original pictures of medicine, and the BD-SW40 biological microscope of Boshi Jingda is used in this embodiment.

[0068] server, including:

[0069] The mask calculation mod...

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 the field of image data processing. Aiming at the problem of insufficient training data, the invention provides a data preprocessing method and system based on image feature refinement, and the method comprises the steps: an effective region calculation step: carrying out mask processing on an original image to obtain a training image, and calculating an effective region of the training image, the effective region comprising a positive sample region and a negative sample region; and a training data acquisition step: constructing a sliding window, moving the position of the sliding window on the training picture, and acquiring training data when the digital center of the sliding window is located at the boundary or inside the effective area. According to the method and system, the problem of insufficient training data acquisition in the existing image processing method can be solved. The invention further discloses a data preprocessing system based on image feature refinement.

Description

technical field [0001] The invention relates to the field of image data processing, in particular to a data preprocessing method and system based on image feature refinement. Background technique [0002] With the development of deep learning becoming more and more popular, more and more deep learning algorithms are applied to the medical industry, and the computer processing of medical images is a relatively active direction. Through computer-aided diagnosis, it can quickly confirm whether there are disease characteristics, reduce the time for doctors to find disease characteristics, and improve the efficiency of disease diagnosis. For example, identifying lesions, judging the degree of cancer, etc. However, medical image data is different from common deep learning data such as faces, cats and dogs, and vehicles. It is affected by the difficulty, complexity, and dimension of medical problems, and the data is extremely scarce. [0003] For the processing of image data, the...

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): G06K9/62G06K9/32
CPCG06V10/25G06F18/24G06F18/214
Inventor 李青海潘宇翔刘翔宇秦于钦赵轩陈钦泽张清瑞赵梦思
Owner GUANGDONG KINGPOINT DATA SCI & TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com