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

Image recognition method, device and device based on depth neural network model

A deep neural network and image recognition technology, applied in biological neural network models, character and pattern recognition, neural architecture, etc.

Active Publication Date: 2019-02-15
湖南极点智能科技有限公司
View PDF8 Cites 23 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the deep neural network has the characteristics of many parameters and a large amount of calculation, resulting in a large amount of memory requirements and computational burden, making it difficult for the deep neural network model to be applied to hardware devices such as mobile phones with limited storage and computing resources.

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 device based on depth neural network model
  • Image recognition method, device and device based on depth neural network model
  • Image recognition method, device and device based on depth neural network model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Please refer to figure 1 , figure 1 It is a flowchart of an image recognition method based on a deep neural network model in an embodiment of the present invention. The method includes the following steps:

[0050] S101. Acquire a target image to be recognized.

[0051] The target image acquisition device is used to collect the target image to be recognized in real time, and the target image to be recognized can also be read in a preset storage device. The target image can be an image to be recognized in real time collected by an image collection device such as a camera.

[0052] S102: Input the target image into the target model obtained by channel pruning the deep neural network model using the representation ability of the channel.

[0053] The deep neural network model can be pruned in advance based on the representation ability of the channel to obtain the target model. Then, after obtaining the target image, the target image can be input into the target model. Since the...

Embodiment 2

[0068] In order to facilitate those skilled in the art to understand the technical solutions provided by the embodiments of the present invention, the auxiliary loss function inserted in the deep neural network model is specifically the cross-entropy loss function as an example, and the auxiliary loss function inserted in the deep neural network model Give details.

[0069] Insert in the deep neural network model Auxiliary loss function { },make{ } Represents the insertion position, where Represents the last layer. Use section Loss function Right Layer for channel selection.

[0070] make Indicates corresponding The input feature map of the input samples, corresponding to the first The definition of a cross-entropy loss function is as follows: ,among them, Represents an exponential function, Indicates the weight of the fully connected layer, Indicates the number of categories, Is the number of input channels of the fully connected layer.

[0071] It should be noted th...

Embodiment 3

[0104] Corresponding to the above method embodiment, the embodiment of the present invention also provides an image recognition device based on a deep neural network model. The image recognition device based on the deep neural network model described below is similar to the image recognition device based on the deep neural network model described above. Image recognition methods can correspond to each other.

[0105] See Figure 4 As shown, the device includes the following modules: a target image acquisition module 101, a target image input module 102, a classification recognition module 103, and a target model acquisition module 104;

[0106] Wherein, the target image acquisition module 101 is used to acquire the target image to be recognized;

[0107] The target image input module 102 is configured to input the target image into the target model obtained by channel pruning the deep neural network model using the representation ability of the channel;

[0108] The classification and...

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 based on a depth neural network model. The target image is inputted into the target model which is obtained by pruning the depth neural network model with the channel representation ability. Using the auxiliary classifier in the target model to classify the target image, the recognition results are obtained. Because the target model of target image recognition is a pruned model based on the channel representation ability, the computational cost of target image recognition can be greatly reduced. The invention also discloses an image recognition device, a device and a readable storage medium based on a depth neural network model, which have corresponding technical effects.

Description

Technical field [0001] The present invention relates to the technical field of intelligent recognition, in particular to an image recognition method, device, equipment and readable storage medium based on a deep neural network model. Background technique [0002] At present, in image classification and recognition such as image classification and face recognition, deep neural network models are very important. However, the deep neural network has the characteristics of many parameters and large amount of calculation, resulting in a large amount of memory requirements and computational burden, making it difficult to apply the deep neural network model to hardware devices such as mobile phones with limited storage and computing resources. [0003] In summary, how to effectively solve the problem of reducing the amount of calculation for image recognition is a technical problem urgently needed to be solved by those skilled in the art. Summary of the invention [0004] The purpose of t...

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): G06K9/62G06N3/04
CPCG06N3/045G06F18/24
Inventor 谭明奎吴希贤
Owner 湖南极点智能科技有限公司
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