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

Face image gender identification model based on convolutional neural network and face image gender identification method based on convolutional neural network

A convolutional neural network and face image technology, applied in biological neural network models, neural learning methods, neural architectures, etc., can solve the problem of low recognition accuracy, achieve enhanced feature responses, simplify backpropagation calculations, The effect of guaranteeing accuracy

Active Publication Date: 2020-04-03
SHANGHAI MARITIME UNIVERSITY
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Experiments show that the convolutional neural network structure effectively overcomes the influence of factors such as illumination and rotation, and has good robustness, but the accuracy of recognition in images with unclear face images and large changes in face poses is not high

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
  • Face image gender identification model based on convolutional neural network and face image gender identification method based on convolutional neural network
  • Face image gender identification model based on convolutional neural network and face image gender identification method based on convolutional neural network
  • Face image gender identification model based on convolutional neural network and face image gender identification method based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] According to the following Figure 2 to Figure 5 , the preferred embodiment of the present invention is described in detail.

[0040] like figure 2 As shown, in one embodiment of the present invention, a convolutional neural network-based face image gender recognition model is provided, including:

[0041] The input layer input is used to input the uniform size face image after preprocessing;

[0042] Deep-CNN convolutional layer deep-CNN, the extracted features contain richer and more complete semantic features;

[0043] Shallow network convolution layer shallow-CNN, the extracted features contain detailed texture edge information;

[0044] The fusion layer concat is used for feature fusion between the features extracted by the deep network convolution layer and the features extracted by the shallow network convolution layer;

[0045] The fully connected layer FC is used to convert the feature map after convolution pooling into a vector;

[0046] The output layer...

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 a face image gender identification method based on a convolutional neural network and a face image gender identification model based on a convolutional neural network. A deep network and a shallow network respectively adopt convolution kernels with different scales to carry out feature extraction on the image to obtain features of different scales and semantics, the calculation amount of the deep network is considered, a Slice layer and an Elasticse layer are added into a network to greatly simplify the model and enhance the feature response, in addition, L-Softmax Lossis introduced as an output layer, the inter-class and intra-class distances are effectively adjusted while the network back propagation calculation amount is simplified, and the calculation amount ofa computer is greatly reduced on the premise that the accuracy is not lost.

Description

technical field [0001] The invention relates to a face image gender recognition method and recognition model based on a convolutional neural network, in particular to a face image gender recognition model and recognition method based on a multi-scale two-way deep and shallow convolutional neural network. Background technique [0002] Gender recognition of face images is an important task of face analysis. Gender recognition is often displayed in identity verification as a necessary attribute in today's face recognition ( figure 1 for common face recognition). Early recognition methods were mostly based on manual feature extraction methods, such as SVM, PCA, and Bayesian decision making. Most of the above methods ignore the two-dimensional correlation of the pixels in the face image, so there is no satisfactory effect on the classification accuracy. With the rise of deep learning, convolutional neural networks are also often used for face gender recognition. The convolution...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V40/172G06V40/168G06N3/045G06F18/213G06F18/253
Inventor 阴紫微陈淑荣
Owner SHANGHAI MARITIME UNIVERSITY