Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Faster RCNN convolutional neural network single-target recognition method based on LeakyRelu activation function

A convolutional neural network and activation function technology, applied in the field of image processing in computer vision, to achieve the effects of expanding applicability, improving detection accuracy, and high detection accuracy

Active Publication Date: 2019-12-13
JILIN UNIV
View PDF6 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention proposes an improved relu activation function, that is, the LeakyRelu activation function is applied to the convolutional neural network to detect and extract the target object in the image, which improves the detection accuracy and improves the existing neural network image processing technology. The convolutional neural network single target recognition method based on the LeakyRelu activation function solves the shortcomings of the current convolutional neural network using relu as the activation function in image recognition and detection

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
  • Faster RCNN convolutional neural network single-target recognition method based on LeakyRelu activation function
  • Faster RCNN convolutional neural network single-target recognition method based on LeakyRelu activation function
  • Faster RCNN convolutional neural network single-target recognition method based on LeakyRelu activation function

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] refer to figure 1 , a convolutional neural network single target recognition method based on the LeakyRelu activation function, the method includes the following steps:

[0015] Step 1. Obtain image samples and data classification and determine the number of training set and test set samples;

[0016] (11) Obtain the image of the front object captured by the camera of the self-driving vehicle:

[0017] Deep learning is a semi-supervised learning algorithm, so enough sample data should be provided in the early stage to input into the convolutional neural network in order to fully learn the characteristics of the picture. In this example, 300 pictures collected from the KITTI data set of roads in a foreign city are collected. The data set includes images in sunny and dark environments, vehicles occluding each other, and roads in complex environments.

[0018] (12) Determine the number of samples in the training set and test set:

[0019] In the collected data set, 60% ...

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 belongs to the field of image processing in computer vision, and particularly relates to a Faster RCNN convolutional neural network single-target recognition method based on a LeakyReluactivation function. The invention provides an improved relu activation function, namely a LeakyRelu activation function, and the function better reserves image data information and is more beneficialto extraction of image features. The function is applied to the field of detection and extraction of a target object in an image in a convolutional neural network, so that the detection accuracy is improved. The detection accuracy of an existing neural network image processing algorithm is improved, and the problems caused by image recognition and detection of a convolutional neural network adopting relu as an activation function at present are solved.

Description

technical field [0001] The invention belongs to the field of image processing in computer vision, and specifically relates to a convolutional neural network single-target recognition method based on a LeakyRelu activation function. Background technique [0002] With the wide application of artificial intelligence technology in various fields of people's life, intelligent driving vehicles have gradually become a major innovation in the automotive field after electric vehicles. In the future, the development direction of the automobile industry is: electrification, intelligence, networking, and sharing. The comprehensive intelligent network connection, new energy, and cloud platform of intelligent driving vehicles have broad research prospects in the automotive field. [0003] In smart driving cars, several sensors are often used instead of human eyes to observe the surrounding external objects and various targets in front. After these sensors replace the human eye to "see" ...

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): G06N3/08G06N3/04
CPCG06N3/082G06N3/045
Inventor 高炳钊范佳琦李鑫
Owner JILIN 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
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