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

Integrated convolutional neural network-based gender recognition method

A convolutional neural network and neural network technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems that affect the performance of face gender recognition, heavy workload, and complex parameter tuning, and achieve a good face. The effect of gender recognition performance, reducing dependencies, and improving accuracy

Active Publication Date: 2017-01-04
SOUTH CHINA UNIV OF TECH
View PDF2 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This process relies heavily on the experience of human experts and repeated experiments. Not only is the workload heavy, but it is also difficult to find an optimal expression of facial gender characteristics, which affects the effect of facial gender recognition.
On the other hand, a single classifier such as a convolutional neural network as a face gender recognition method needs to adjust many parameters, parameter tuning is complicated, the workload is heavy, and it is difficult to obtain optimal parameters, thus affecting the overall face gender recognition performance

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
  • Integrated convolutional neural network-based gender recognition method
  • Integrated convolutional neural network-based gender recognition method
  • Integrated convolutional neural network-based gender recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0050] This embodiment discloses a gender recognition method based on an integrated convolutional neural network, such as figure 1 As shown, the steps are as follows:

[0051] S1. First, a number of new training data sets are randomly combined to form several new training data sets, and then M convolutional neural network classifiers obtained through the training of the above new training data sets are selected as base classifiers, which are respectively the first base classifier, the second base classifier Base classifier, ..., the Mth base classifier; such as figure 2 As shown, the process of obtaining the base classifier in this step is as follows:

[0052] S11. Select a benchmark data set and several auxiliary data sets; wherein the benchmark data set is divided into a benchmark training data set and a benchmark test data set; as image 3 As shown, in the present embodiment, the Feret data set is selected as the benchmark data set, and the Adience data set and the AR da...

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 integrated convolutional neural network-based gender recognition method. The method comprises the following steps: S1, first carrying out random combination to form a plurality of new training data sets, and selecting M convolutional neural network classifiers which are obtained through training by using the abovementioned new training data sets to serve as base classifiers; S2, obtaining a to-be-tested facial image; and S3, during the test, respectively inputting the to-be-tested facial image into the M base classifiers obtained in the step S1, and fusing gender types output by the M base classifiers to obtain a final gender type. According to the method disclosed by the invention, the convolutional neural network classifiers which are obtained through training by using the randomly obtained new training data sets serve as the base classifiers, the to-be-tested facial image is input into the M base classifiers, and finally the gender types output by the M base classifiers are fused to obtain the final gender type, so that the method has the advantages of being high in recognition accuracy rate, reducing the dependency, on the people, of the gender feature extraction of the facial image, and being wide in application.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a gender recognition method based on an integrated convolutional neural network. Background technique [0002] Face gender recognition has great application value in personalized services, intelligent monitoring, etc. For example, in product recommendation, the product needs of men and women are very different. Facial gender recognition will help merchants understand the different preferences and needs of men and women, so as to improve products and provide better user experience and content quality. Face gender recognition can also help smart devices such as wearable devices to provide customized services, because different genders have different preferences, living habits and unique characteristics. [0003] At present, there are many face gender recognition methods, and the gender recognition system is generally divided into two parts: feature extraction of face imag...

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): G06K9/00G06K9/62
CPCG06V40/172G06F18/2415
Inventor 文贵华吴泽银
Owner SOUTH CHINA UNIV OF TECH
Features
  • Generate Ideas
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More