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

Texture segmentation algorithm based on shape descriptor and twin neural network

A technology of shape description and neural network, applied in computing, image analysis, image data processing, etc., can solve problems such as grouping descriptor ambiguity, segmentation error, etc., and achieve good segmentation effect and good texture segmentation effect

Active Publication Date: 2020-02-14
铁道警察学院
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this leads to ambiguity in grouping descriptors, especially for nearby descriptor boundaries
This can lead to segmentation errors if the descriptors are grouped to form segmentations, and this problem is exacerbated when the texture geometry in the image is large

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
  • Texture segmentation algorithm based on shape descriptor and twin neural network
  • Texture segmentation algorithm based on shape descriptor and twin neural network
  • Texture segmentation algorithm based on shape descriptor and twin neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0093] In the Siamese neural network structure, two symmetrical networks each have a fully connected layer. The input base shape descriptor is a 40-dimensional descriptor, that is, RGB channels, grayscale and 4 directional gradients on 5 scales, where the 5 scales are α=(10,20,30,40,50) . The output descriptor f of the Siamese network is the same as the number of hidden units used. The sigmoid function of the weighted difference of the two Siamese networks is used to compute the metric D for a pair of descriptors.

[0094] In order to compare the texture segmentation results, in this embodiment, the texture segmentation algorithm of the present invention is compared with the mcg, gPb, CTF, and STLD algorithms. Then, use ODS and OIS indicators to evaluate the algorithm's boundary accuracy and area accuracy. For all metrics, higher values ​​indicate that the segmentation results are closer to the ground truth.

[0095] In this embodiment, the Brodatz synthetic data set is us...

Embodiment 2

[0100] Experiment with real texture datasets. Use the literature "N.Khan, M.Algarni, A. Yezzi, and G. Sundaramoorthi. Shape-tailored local descriptors and their application tosegmentation and tracking. In Proceedings of the IEEE Conference on ComputerVision and Pattern Recognition, pages 3890–3899, 2015" The 128 images in the dataset and the 150 images in the Berkeley segmentation dataset are used as the training set, and the remaining images are used as the test set. Then, the algorithm proposed by the present invention is initialized by 5×5 standard block subdivision. The final segmentation result indicators are shown in Table 2. From Table 2, it can be seen that the texture segmentation algorithm of the present invention has the best effect on both the contour index and the area index.

[0101] Table 2 Results of Texture Segmentation Dataset

[0102]

[0103] Among them, a schematic diagram of the experimental results is attached Figure 9-15 shown. by attaching F...

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 a texture segmentation algorithm based on a shape descriptor and a twin neural network, and the algorithm comprises the steps: assuming that an image consists of regions, and each region is provided with a fixed shape descriptor; combining each shape descriptor with a twin neural network; the texture segmentation is then designed into an optimization problem, when a regionof interest is selected, the segmentation is an optimal solution, so that the learned shape descriptor is almost constant in the region, and the optimal solution is solved according to a method of minimizing energy. The texture segmentation algorithm provided by the invention is superior to other algorithms in terms of contour indexes and regional indexes, and can achieve a better segmentation effect on complex geometric transformation or complex nuisance images.

Description

technical field [0001] The invention relates to texture segmentation algorithms, in particular to a texture segmentation algorithm based on shape descriptors and twin neural networks. Background technique [0002] Image texture can be qualitatively described by physical quantities such as intensity, density, and direction. Texture segmentation of images is a fundamental problem in the field of computer vision, and its segmentation quality plays a key role in image post-processing tasks such as object classification and extraction. At present, image texture segmentation has gradually become an important research direction in the field of image analysis. [0003] Common methods for texture segmentation can be roughly divided into two categories, namely edge-based methods and region-based methods. Among them, edge-based methods attempt to localize edges as responses to filter banks, and then post-process these responses to fill gaps and generate segmentations. Although edge-...

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): G06T7/11
CPCG06T7/11G06T2207/20084G06T2207/20081
Inventor 李卫平武海燕
Owner 铁道警察学院
Features
  • Generate Ideas
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com