Intelligent terminal human-computer interaction method capable of fusing human face and gesture

Abstract

The invention discloses an intelligent terminal human-computer interaction method capable of fusing a human face and a gesture, and belongs to the image processing technical field. The intelligent terminal human-computer interaction method capable of fusing the human face and the gesture includes: step one, performing normalization processing on human face images obtained by an intelligent terminal camera; step two, calculating center coordinates of major characteristic of the human face after the normalization processing is performed on the obtained human face images; step three, performing hand complexion and background modeling according to gesture images collected by the intelligent terminal camera; step four, describing characters of the gesture images detected in the step three.

Description

technical field [0001] The invention relates to a human-computer interaction method for an intelligent terminal that integrates human faces and gestures, and belongs to the technical field of image processing. Background technique [0002] With the rapid popularization of smart phones, smart TVs, tablet computers, and vehicle-mounted intelligent systems, the interaction between people and smart terminals has become an important part of people's daily life. In this case, more natural and convenient human-computer interaction technologies based on voice, face, and gestures have developed rapidly. [0003] Face recognition technology provides a natural method of human-computer interaction. Human face is the most common pattern in human vision, and it is the easiest to be accepted by humans. The visual information represented by the human face plays an important role and significance in social communication. Face recognition technology relies on various disciplines such as pa...

AI Technical Summary

Problems solved by technology

[0005] To sum up, whether it is face or gesture, its recognition effect is not ideal at present.

Studies have found that a single biological feature (such as face, hand shape, fingerprint, etc.) contains limited information, which is one of the main reasons why it cannot meet people's diverse needs (naturalness, convenience, security, etc.)

Literature [5] reveals that the commonly used hand shapes and faces contain only 10 5 and 10 3 order of magnitude, so it is impossible for a system based on a single face or hand shape recognition technology to meet the diverse needs of people's daily life

In addition, although fingerprints and irises have a large amount of information content, the existing fingerprint authentication systems are not optimistic about the recognition rate of low-quality fingerprint images.

Therefore, for some key applications, it is difficult for a single biometric system to meet people's needs

Images