Human eye sight line detection method and device based on photogrammetry technology

Abstract

The invention discloses a human eye sight line detection method and device based on the photogrammetry technology. The method comprises a step of setting a stereo target to calibrate an image acquisition unit and obtaining internal parameters and / or external parameters of the image acquisition unit, continuously acquiring human face images in the time sequence by using the image acquisition unit and processing each of the human face images and obtaining iris center pixel coordinates of human eyes, carrying out three-dimensional reconstruction based on the internal parameters and / or external parameters of the image acquisition unit and the iris center pixel coordinates to obtain human eye three-dimensional coordinates and saving the human eye three-dimensional coordinates according to the time sequence, and reconstructing a human eye motion trajectory according to the human eye three-dimensional coordinates saved in the time sequence. The human eye sight line detection method and devicebased on the photogrammetry technology provided by the invention can be applied to provide data information for visual fatigue assessment, and hidden dangers caused by visual fatigue on human healthand traffic safety are avoided.

Description

technical field [0001] The invention relates to a photogrammetry method, in particular to a photogrammetry technology-based human sight line detection method. The invention also relates to a human eye sight detection device. The human eye sight detection method and device based on photogrammetry technology provided by the present invention can be applied to provide data information for visual fatigue assessment, and avoid hidden dangers caused by visual fatigue to human health and traffic safety. Background technique [0002] Photogrammetry technology is widely used in shape detection, low-frequency vibration measurement, and defect detection. Three-dimensional visual inspection technology has the advantages of fast, non-contact, real-time and so on. Photogrammetry technology combined with image processing technology can extract the part of interest in the image for motion trajectory detection. There are relatively mature applications for the motion trajectory detection o...

AI Technical Summary

Problems solved by technology

This method extracts the pupil size and blink frequency, which can reflect the driver's fatigue to a certain extent, but this method does not detect the human line of sight, and cannot guarantee the reliability of the detection.

[0008] It can be seen that although the detection based on human vision in the prior art can be used to a certain extent to detect the driver's fatigue when applied to vehicle driving, the technical solution itself does not It is not fully combined with the specific application environment, so there are actual defects of low reliability

Images