A head-mounted visual precision aiming system

Abstract

The present invention provides a head-mounted visual precise aiming system, which integrates visual aiming control technology with a deep learning-based target detection method, and can perform precise visual aiming. The method is as follows: the human eye detection and gaze point output module obtains the real-time gaze point coordinates and sends them to the data fusion processing module; the deep learning target detection module detects the target in the field of view scene image based on the deep learning target detection model obtained through training Real-time detection, and the detected target is marked in the field of view scene image in the form of a candidate target frame and then sent to the head-mounted device; at the same time, the bounding box information of each candidate target is sent to the data fusion processing module; the data fusion processing module is based on The coordinates of the gaze point and the bounding box information of each candidate target are received, and the position coordinates of the selected target and the aiming and tracking servo control amount for tracking the selected target are obtained, and the servo pan / tilt is controlled to perform aiming and tracking on the selected target.

Description

technical field [0001] The invention relates to an aiming system, in particular to a system capable of precise visual aiming, and belongs to the technical field of visual aiming of head-mounted devices. Background technique [0002] At present, about 44 eye tracker research and development and production units in the world are developing head-mounted eye movement detection equipment. Among them, the more mature ones include the American ASL Applied Science Laboratory, the Canadian SR Scientific Research Institute, the German SMI Sensing Motion Instrument Company and Sweden's Tobii eye tracker company. These eye movement detection devices can basically collect the corresponding data of the human eye, and analyze the collected data and apply it to related research fields. Domestic related research fields have introduced some desktop and head-mounted eye-tracking devices to further develop their own eye-movement measurement and gaze tracking systems. These eye-tracking device...

AI Technical Summary

Problems solved by technology

[0006] (1) Due to the physiological characteristics of the human eye, in the process of gazing, the human eye trembles, etc., the fixation point data solved also has certain fluctuations, which affects the actual fixation accuracy;

[0007] (2) The actual gaze accuracy of the existing eye-tracking equipment can only reach 0.5°~1°, or even greater than 1°. It is difficult to apply the gaze point data to the servo control of the pan / tilt. Realize the precise control of the servo motion of the pan / tilt;

[0008] (3) The target detection method based on deep learning can accurately detect the target, but for the detection of multiple targets of the same type in the scene, it is not possible to selectively and distinguishably detect a specific target of the same type

Images