Face image age judgment method and device, computer equipment and memory medium

Abstract

The embodiment of the invention discloses a face image age judgment method and device, computer equipment and a memory medium. The method comprises the following steps of obtaining to-be-judged face images; inputting the face images into a preset neural network model; and carrying out age classification on the face image according to classification data output by the neural network model. When theneural network model for carrying out face age judgment is trained, uniform expectation processing is carried out on the images of the same person photographed in different space environments, so a plurality of age expectations of different images of the same person are obtained; the age expectations are ranked and a median in a ranking result is taken out; and the age expectations of the different images of the same person in a training sample set are the same, so according to the neural network model trained and converged through utilization of the images, the age score output stability forthe same person in the different environments is high, and an age score is not liable to be influenced by the environments.

Description

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

AI Technical Summary

Benefits of technology

This technology allows users trained on their own computer networks (the Neural Network) to predict how likely someone will be looking at something differently based upon factors like location or lighting conditions. It uses an ensemble method called Uniform Thermal Processor Analysis ("UTPA") which involves analyzing images from various locations around them under specific environmental settings such as temperature. By comparing these temperatures against predetermined thresholds, it can determine if there was any difference between two scenarios - one where individuals were facing outdoor sunlight while another scenario had indoor lights illuminated during daytime hours. These differences indicate who would look up too much over time compared to what people actually see outside.

Problems solved by technology

This patented technical problem addressed in the current study relates to improving the performance of face identification systems used on faces like cars due to their ability to accurately recognize them over time. However, previous methods were limited because they resulted in similar appearances between identically captured photos compared against each other's versions.

Images