Video-based pedestrian gender and age recognition method and model

Abstract

The present invention provides a video-based pedestrian gender and age recognition method and model, the method includes acquiring an image to be tested; performing human body contour detection, identifying pedestrians and using a human body area detection frame to frame, cutting out the human body contour picture for key point extraction, Select multiple human body contour pictures of pedestrians, splicing pixel value coordinates to output a time-series initial vector, normalize to obtain a time-series normalized vector, and align the human body contour pictures to obtain a human body contour alignment image; horizontally cut out multiple human body partial contour sub-pictures After deep feature extraction, output the spatial initial vector containing global feature information and local feature information, and normalize to obtain the spatial feature vector; obtain the decision vector to obtain the gender and age judgment results. The invention solves the problem of low privacy and security due to the need to extract and collect face information data of pedestrians when image recognition processing technology is used to acquire pedestrian's gender and age information in the prior art.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence computer vision recognition, in particular to a video-based pedestrian gender and age recognition method and model. Background technique [0002] With the rapid development of artificial intelligence technology, the application of computer vision recognition technology is constantly popularizing, and it has played an important role in the fields of security prevention, transportation, business operations and even criminal investigation. Computer vision recognition technology mainly uses a computer-built neural network model that simulates the human nervous system to recognize and process images to obtain the required information data and replace humans to efficiently complete specific tasks in a certain field. [0003] In the application of computer vision recognition technology, an important information data obtained by using it is the gender and age of pedestrians. In the prior ...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of the present invention is to provide a video-based pedestrian gender and age recognition method and model to solve the need to extract and collect pedestrian facial information data when using image recognition processing technology to obtain pedestrian gender and age information in the prior art , so there is a risk that the facial information data of pedestrians will be stolen intentionally or leaked unintentionally, which has certain potential safety hazards

Images