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

A nanoparticle size measurement method based on an improved Mask R-CNN

A nanoparticle and measurement method technology, applied in the field of deep learning and image processing, can solve problems such as poor robustness, difficult design, and complicated steps, and achieve the effects of accelerating convergence, improving classification accuracy, and improving training speed

Inactive Publication Date: 2019-06-28
TIANJIN POLYTECHNIC UNIV
View PDF4 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional image segmentation algorithms need to manually design feature extraction methods, which are not only difficult to design, complicated steps, but also poor robustness

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
  • A nanoparticle size measurement method based on an improved Mask R-CNN
  • A nanoparticle size measurement method based on an improved Mask R-CNN
  • A nanoparticle size measurement method based on an improved Mask R-CNN

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The present invention will be further described in detail below in conjunction with specific embodiments.

[0031] A schematic diagram of the overall framework of the present invention is as figure 1 As shown, first of all, since there is no public nanoparticle image library, it is necessary to collect spherical and rod-shaped nanoparticle images to make a data set, use the LabelMe graphic annotation tool to manually mark the particle area of ​​the training sample; introduce the DenseNet structure, the normalization layer And Arcface Loss loss function to improve the overall framework of the Mask R-CNN network, use the validation set to adjust the parameters of the improved Mask R-CNN, and use the improved MaskR-CNN to test the test set; to fit the nanoparticle boundary , Adopt different fitting methods for different nanoparticles; measure the particle size parameters of the particles, and use the diameter of the round edge obtained by fitting spherical nanoparticles as the...

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 provides a Mask R-based on improvement. The invention discloses a CNN nanoparticle size measurement method. The method comprises the following steps: 1) collecting spherical and rodlikenanoparticle images to make a data set; 2) improving Mask R- CNN and training, using improved Mask R- The CNN is used for segmenting the nanoparticles; 3) performing pixel calibration to obtain an actual size corresponding to each pixel in the image; 4) fitting the nano-particle boundary; And 5) measuring particle size parameters of the particles. According to the method, a deep learning technology is utilized, a model can be trained under the condition of a large sample, particle characteristics in an image can be automatically extracted, a complex target segmentation problem is converted into weight parameters, the steps are simple, the robustness is high, the problem that a traditional method is inaccurate in segmentation is solved, and the accuracy of nano particle measurement is improved.

Description

Technical field [0001] The invention relates to a nano particle size measurement method based on an improved Mask R-CNN, which is used for the measurement of nano particle size and is more accurate than traditional algorithms in terms of accuracy and belongs to the field of image processing and deep learning. Background technique [0002] Nanotechnology is widely used in industries such as catalysis science, medical drugs, new materials, power industry and composite materials, and has an important position in the entire high-tech field. Since many characteristics of nanomaterials have an important relationship with their particle size, shape and other microstructures, the characterization of the microstructure of nanomaterials is important for understanding the characteristics of nanomaterials, seeking application fields of nanomaterials, and promoting the development of nanomaterials. An important role, and the size measurement of nanoparticles is the key technology. At present...

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): G06K9/62G06K9/46G06T7/12G01N15/02
Inventor 张芳张乾肖志涛耿磊吴骏刘彦北王雯吴玥
Owner TIANJIN POLYTECHNIC UNIV
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