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

Channel logo segmentation method for pixel-level channel logo recognition network based on cross-layer feature extraction

A technology for station logo recognition and feature extraction, applied in image analysis, neural learning methods, character and pattern recognition, etc. It can solve problems such as limited performance, limited algorithm performance, training samples, and poor recognition results, and achieve strong description. Ability and discrimination ability, strong description ability, and the effect of improving output accuracy

Active Publication Date: 2018-05-08
TIANJIN UNIV
View PDF7 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In summary, due to the limited description ability of manual features, the recognition effect is not good
Existing logo recognition methods based on machine learning such as support vector machines or artificial neural networks are limited by the performance of feature extraction algorithms. The network pays more attention to semantic features and cannot extract fine local features of logos.
At the same time, the performance improvement of these algorithms is also limited by the number of training samples

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
  • Channel logo segmentation method for pixel-level channel logo recognition network based on cross-layer feature extraction
  • Channel logo segmentation method for pixel-level channel logo recognition network based on cross-layer feature extraction
  • Channel logo segmentation method for pixel-level channel logo recognition network based on cross-layer feature extraction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] A method for segmenting a station logo based on a pixel-level station logo recognition network based on cross-layer feature extraction of the present invention will be described in detail below with reference to the embodiments and the accompanying drawings.

[0047] Such as figure 1 As shown in the present invention, a method for segmenting a station logo based on a pixel-level station logo recognition network based on cross-layer feature extraction includes the following steps:

[0048] 1) Based on the existing classification network, modify the fully connected layer to a convolutional layer, and introduce a cross-layer architecture. By connecting the low and high layers of the classification network, and combining the characteristics of the input image output by the low and high layers, extract Combining cross-layer features of image local features and global features, three pixel-level logo recognition networks (PNET) with different cross-layer architectures are res...

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

A channel logo segmentation method for a pixel-level channel logo recognition network based on cross-layer feature extraction includes the steps of modifying a fully connected layer of an existing classification network to a convolutional layer, combining features of input images output by low and high layers, extracting cross-layer features that integrate local and global features of the images,and constructing three pixel-level channel log recognition networks of different cross-layer architectures; using existing channel logo data sets as training and test data, including a channel logo image set and a binary label image set corresponding to the channel logo image set, and extracting three cross-layer features of the channel logo image set respectively by using the three pixel-level channel logo recognition networks; training the pixel-level channel logo recognition networks; and testing the channel logo image set by using the three pixel-level channel logo recognition networks trained on the channel logo data set of different types respectively, and finally producing pixel-level segmentation results of the same size as the input images. The channel logo segmentation method ofthe invention has a stronger description capability and discrimination capability.

Description

technical field [0001] The invention relates to a station logo segmentation method. In particular, it relates to a station logo segmentation method of a pixel-level station logo recognition network based on cross-layer feature extraction. Background technique [0002] The feature extraction of the logo image is the key factor to determine the accuracy of the logo recognition. Finding a more distinguishing feature has always been the goal pursued by researchers. At present, logo recognition mainly adopts the method of extracting features first and then calculating feature similarity. The features used include traditional features and deep features. [0003] The features used in traditional station logo recognition methods can be divided into global features and local features. Among them, global features usually include features such as color, shape, mesh, principal component analysis, edge, and texture. Although the color information has the characteristics of size and vi...

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): G06N3/08G06N3/04G06T7/194G06K9/62G06K9/46
CPCG06N3/08G06T7/194G06V10/44G06N3/045G06F18/24
Inventor 张静徐佳宇苏育挺张冬明
Owner TIANJIN 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