Unlock instant, AI-driven research and patent intelligence for your innovation.

Image recognition method and device based on lightweight deep neural network

A deep neural network and image recognition technology, which is applied in the field of image recognition and devices based on lightweight deep neural networks, can solve problems such as high requirements and image classification, and achieve the effect of improving the speed of image recognition

Pending Publication Date: 2020-09-04
XIANGTAN UNIV
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, deep neural networks have high requirements for resources and space, and it is impossible to apply deep neural networks to classify images on some resource-constrained platforms.

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 and device based on lightweight deep neural network
  • Image recognition method and device based on lightweight deep neural network
  • Image recognition method and device based on lightweight deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] This embodiment provides an image recognition method based on a lightweight deep neural network, such as figure 1 shown, including the following steps:

[0055] Step 1. Construct a deep neural network model for image recognition; train the constructed model based on the training set to obtain a trained model;

[0056] Step 2. Lightweight network model: Initialize the pruning parameters (parameters of reinforcement learning), including the pruning step size of each feature layer (the pruning step size of each feature layer can be different), and cyclically update the model parameters and pruning parameters; The process of each cycle is as follows:

[0057] First, for the current feature layer in the current model M, use any method to evaluate the importance of channels to sort the importance of each channel, and prune the α channels with lower importance to obtain the model after this round of pruning M'; where the feature layer refers to the module consisting of a con...

Embodiment 2

[0064] In this embodiment, on the basis of Embodiment 1, in the step 2, for any feature layer in the current model M, the importance degree of each channel is calculated based on the scaling coefficient corresponding to each channel in the batch normalization layer. Sorting; the channel with the larger corresponding scaling factor is more important. The channel importance is sorted based on the scaling factor, which greatly improves the sorting speed compared to the sorting based on the feature map.

Embodiment 3

[0066] This embodiment is based on the embodiment 1, in the step 2, for the less important α i Pruning the channels refers to setting the weights associated with the α channels in the convolutional layer, the fully connected layer, and the normalization layer to 0; in addition to resetting the weights to 0, model reconstruction can also be used, that is, Remove the α in the original model i A method of channel-associated structure for model pruning.

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 and device based on a lightweight deep neural network. The method comprises the following steps: 1) constructing and training a deep neural networkmodel for image recognition; 2) a lightweight model: updating parameters and trimming parameters of the model circularly, and each round of circulation process is as follows: firstly, trimming channels with relatively low importance in a current feature layer of the current model M to obtain a trimmed model M 'of the current round, then training the trimmed model M' again, and calculating the relative precision of the trained model M '; if the relative precision is not negative, calculating the award of the current round of pruning, updating the pruning parameters, and then carrying out the next round of circulation; if the relative precision is negative, abandoning M ', returning to M, and determining whether to perform the next round of circulation or end the circulation after the trimming parameter is updated according to whether the precision loss is within an allowable precision loss range; and 3) recognizing the to-recognized image by means of the final model; the image recognition method based on the resource-constrained platform is suitable for image recognition of the resource-constrained platform.

Description

technical field [0001] The present invention is an image recognition method (classification) method and device based on a lightweight deep neural network. Background technique [0002] Since the deep neural network was proposed, it has been widely used in the field of image recognition, and it has good effect on image recognition and high accuracy. However, deep neural networks have high requirements for resources and space, and deep neural networks cannot be used to classify images on some resource-constrained platforms. [0003] To address this problem, it is necessary to provide a method for image recognition based on a lightweight deep neural network on a resource-constrained platform. Contents of the invention [0004] The technical problem solved by the present invention is to provide an image recognition method and device based on a lightweight deep neural network, which can perform image recognition based on a deep neural network on a platform with limited resourc...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/082G06N3/045G06F18/241G06F18/214
Inventor 王冬丽刘广毅周彦
Owner XIANGTAN UNIV