A visual attention detection method based on an improved mixed increment dynamic Bayesian network

Abstract

The invention requests to protect a visual attention detection method based on an improved mixed increment dynamic Bayesian network. According to the method, a head, a sight line and a prediction sub-model are fused to carry out comprehensive estimation on the visual attention detection method; the sight line detection sub-model is improved on the basis of a traditional human eye model, so that the recognition rate is increased, and the detection robustness of different testers is improved; in order to solve the problem of data loss caused by extreme postures and dynamic scenes, a prediction sub-model is provided, and the correlation of sampling pictures at two moments is measured by utilizing a Gaussian covariance, so that the false recognition at the current moment is effectively improved, and the recognition error is reduced. Secondly, describing the related sub-models, and respectively establishing Bayesian regression models by utilizing conditional probabilities; And parameters ofthe model are dynamically updated by using an incremental learning method, so that the adaptability of the whole model to new input data is improved.

Description

technical field [0001] The invention belongs to the field of image processing and pattern recognition, in particular to a visual attention detection method based on an improved hybrid incremental dynamic Bayesian network. Background technique [0002] Visual focus of attention (VFOA) specifically refers to the direction and target that the human eye pays attention to. It represents the direction a person is looking at, and contains rich information, such as: what the person is interested in, what he is doing, etc. This information has high application value for fields such as human-computer interaction, intelligent assisted driving, medical research, psychology, and market analysis. In recent years, especially in terms of human-computer interaction, visual attention, as an input method to assist other commands, and judge whether it is a user or other human obstacles in front of it, so as to realize intelligent bionic obstacle avoidance, has attracted extensive attention fro...

AI Technical Summary

Problems solved by technology

[0003] At present, there are two main problems in the detection of visual attention in dynamic scenes and free head postures that need to be studied: under normal eye use, the head posture deflection is consistent with the direction of the line of sight, but the situation that the direction is inconsistent cannot be ruled out, so only the head posture is used. The recognition rate of visual attention is low and the reliability is not high; the impact of low image resolution

When performing line of sight detection, when the resolution is low, the width of the eye is only a few tens of pixels, and the iris area is only a few to a dozen pixels, so it is difficult to reliably detect the continuous iris outline and the position of the corner of the eye, resulting in a deviation in the estimated result of the line of sight direction

Attention detection based on gaze estimation is less robust to dynamic changes (free head rotation, distance changes)

Images