Deep neural network method based on information lossless pooling

A neural network and pooling technology, applied in biological neural network models, neural architectures, instruments, etc., can solve the problems of high complexity of pooling operations and loss of feature information, so as to improve network performance, improve performance, and achieve simple Effect

Inactive Publication Date: 2017-12-15
TIANJIN UNIV
View PDF3 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0013] The purpose of the present invention is to overcome the problems of feature information loss and high complexity of pooling operations in the existing deep neural network pooling layer operation, and propose a deep neural network

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
  • Deep neural network method based on information lossless pooling
  • Deep neural network method based on information lossless pooling
  • Deep neural network method based on information lossless pooling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] Below in conjunction with accompanying drawing this patent is described further.

[0035] figure 1 (a)(b) describe traditional pooling operations. In the traditional pooling operation, assuming that a neighborhood contains four values ​​(1.5, 1.1, 2.0, 0.8) as shown in Figure (a), after traditional pooling operations, such as the maximum pooling operation, the output is the largest A value of 2.0 is used as the output of the current neighborhood. That is, a numerical value is used to replace the current neighborhood value, while other values ​​are discarded. As shown in (b), after the traditional pooling operation (step size is 2), the dimension of the single feature map is reduced to half of the original. However, in this process, part of the information is lost and cannot be recovered, which is an information-damaging pooling operation, which limits the performance of the neural network when it is applied to tasks such as image recognition.

[0036] figure 1(c)(d...

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 relates to a deep convolutional neural network method based on information lossless pooling. The method is used for image classification, and comprises the steps of collecting different kinds of images and marking image categories as image label information; performing image set division, and dividing the collected images into a training set, a verification set and a test set; designing a convolutional neural network structure based on information lossless pooling, including the number of convolutional layers and the number of information lossless pooling layers that are used, designing the number of filters in the convolutional layers, designing Gaussian smoothing filter parameters of the information lossless pooling layers, the pooling window size and a convolutional filter structure for feature fusion, designing the number of network training loop iterations and network final convergence conditions, and initializing network parameters; and inputting training data to the network in batches for calculation and training.

Description

technical field [0001] The invention relates to a method for high-performance picture classification and object recognition in the field of computer vision, in particular to a method for carrying out picture classification and object recognition using a deep learning method. Background technique [0002] In recent years, deep learning technology has been widely used in multiple tasks such as image classification, semantic segmentation and object detection, and automatic driving in the field of computer vision. As an important implementation method in deep learning technology, deep convolutional neural network has achieved remarkable results in many tasks. [0003] Deep convolutional neural networks are often composed of multi-layer convolutional layers and pooling layers. The convolutional layer contains filter parameters for feature extraction, and the pooling layer is used to maintain the translation invariance of the neural network and reduce the impact of data disturbanc...

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): G06K9/62G06N3/04
CPCG06N3/045G06F18/24G06F18/214
Inventor 李亚钊庞彦伟
Owner TIANJIN 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