Capsule-Net-based face expression recognition method

Abstract

The invention discloses a Capsule-Net-based face expression recognition method. The method comprises: step one, selecting a data set and selecting a Cohn-Kanade Plus Database expression data set, thatis, a CK+ expression data set; step two, selecting a deep learning model and using a deconvolution-version Capsule Net as an experimental model; step three, carrying out training and testing based onthe model, segmenting the CK+ expression data set into a training data set and a test data set, reading the training data set into the Capsule Net for periodic training to obtain a stable and accurate Capsule Net deep learning model; and step four, combining the trained Capsule Net deep learning model with an NAO robot, tracking the training process in real time by the NAO robot, reporting information related with the model testing regularly, and expressing an image recognition result in a voice and gesture manner. Therefore, convergence is realized quickly within short time; and the facial expression recognition accuracy and the reliability are improved.

Description

technical field [0001] The invention relates to the technical field of image processing and classification, in particular to a Capsule Net-based facial expression recognition method. Background technique [0002] Image classification is one of the research topics in the field of artificial intelligence, that is, to label the input image with a fixed category, which is one of the core issues in the field of computer vision. It has different practical applications. It is used in industry to detect, identify and classify products, which accelerates the process of intelligent industrialization. It is used in the medical field for disease diagnosis and treatment. It can also be used in robotics to process and classify images. The combination of technology and robots makes robots humanized and intelligent. [0003] Specifically, classic machine learning algorithms can be used to classify image datasets. The powerful feature extraction ability and good classification effect of th...

AI Technical Summary

Problems solved by technology

However, these classic deep learning models only have geometric invariance in a small range of images, but cannot dig out the spatial relationship between features in a large range. In an image, the nose and eyes are the relationship between the eyes and the nose. , if the position is reversed, it cannot be regarded as an image of a human face, and the traditional deep convolutional neural network will still recognize it as a human face

Images