Image recognition method, device and equipment based on lightweight network model

A network model and image recognition technology, applied in the field of pattern recognition, can solve the problems of low accuracy of wide network, dependence on expensive hardware configuration and time-consuming calculation and training

Active Publication Date: 2019-09-27
CHONGQING UNIV
View PDF10 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to overcome the above-mentioned deficiencies in the prior art, and provide an image recognition method, device, equipment and readable storage medium based on a lightweight network model, so as to realize the size, efficiency, resources, and accu

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 recognition method, device and equipment based on lightweight network model
  • Image recognition method, device and equipment based on lightweight network model
  • Image recognition method, device and equipment based on lightweight network model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0071] Such as Figure 1 to Figure 3 As shown, this embodiment provides an image recognition method based on a lightweight network model, which specifically includes the following steps:

[0072] S1, acquiring a target image to be identified;

[0073] S2, inputting the target image into the trained lightweight network model;

[0074] S3, using the trained lightweight network model to classify the target image.

[0075] Wherein, the process of obtaining the lightweight network model includes the following steps:

[0076] S21, according to the construction method of the convolutional neural network, construct a variant convolutional neural network (CNN) without a fully connected layer, the variant convolutional neural network includes one or more network layers, and the network layer includes a convolutional layers and a pooling layer;

[0077] S22, input the marked image to be classified; the marked image to be classified obtains the feature of the image through the variant...

Embodiment 2

[0142] The image recognition method based on the lightweight network model described in Embodiment 1 can be widely used in the field of image recognition. In this embodiment, the MNIST data set is used to test and train the network. The detailed process is as follows:

[0143] Construct a variant convolutional neural network (CNN) to extract image features. Construct a variant CNN based on the LeNet network structure. In the variant convolutional neural network, the first convolutional layer, the first pooling layer, the second convolutional layer, and the second pooling layer are connected in sequence, and the first convolutional layer The size of the convolution filter and the second convolutional layer are both 5*5, and the moving step is 1, where the depth of the filter of the first convolutional layer is 32, and the depth of the filter of the second convolutional layer is 64; the first pooling layer and the second pooling layer both use the maximum pooling method, and the...

Embodiment 3

[0174] Corresponding to the above method embodiment, this embodiment also provides an image recognition device based on a lightweight network model, the image recognition device based on a lightweight network model described below and the image recognition device based on a lightweight network model described above The identification methods can be referred to in correspondence with each other.

[0175] see Figure 5 As shown, the device includes the following modules: target image acquisition module 101, target image input module 102, classification recognition module 103 and target model acquisition module 104;

[0176] Wherein, the target image acquisition module 101 is used to acquire the target image to be identified;

[0177] The target image input module 102 is used to input the target image into the target lightweight network model; the lightweight network model is obtained by removing the variant convolutional neural network and the width network structure of the ful...

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 recognition method, device and equipment based on a lightweight network model. The image recognition method comprises the following steps: S1, obtaining a to-be-recognized target image; s2, inputting the target image into a trained lightweight network model; and S3, classifying the target image by using the trained lightweight network model, wherein the process of obtaining the lightweight network model comprises the following steps: S21, constructing a variant convolutional neural network without a full connection layer; s22, classifying the images through a softmax classifier, and updating the weight of the convolutional layer; s23, extracting features of the image again by adopting the variant convolutional neural network with updated weight, and carrying out standardized processing on the features; and S24, generating feature nodes and enhancement nodes from the standardized features according to a construction method of a width network, determining the final feature nodes and the number of the enhancement nodes, and constructing a lightweight network model.

Description

technical field [0001] The invention relates to the technical field of pattern recognition, in particular to an image recognition method, device, equipment and readable storage medium based on a lightweight network model. Background technique [0002] When the deep neural network is applied to the field of image recognition, since the deep neural network involves a large number of hyperparameters and complex structures, this complexity makes it very difficult to analyze the deep structure theoretically. Most of the work involves adjusting parameters or stacking more layers to Obtain better accuracy, so although the deep neural network has high accuracy, it takes a long time to calculate and train. The width network (BLS) proposed in the article "Broad Learning System: An Effective and Efficient Incremental Learning System Without the Need for Deep Architecture" is designed based on the idea of ​​RVFLNN. Compared with the "depth" structure, the "width" structure has no layers...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/241Y02T10/40
Inventor 房斌李婷
Owner CHONGQING UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap