Posture resolving method based on multi-sensor fusion

Abstract

The invention discloses a posture resolving method based on multi-sensor fusion, and belongs to the field of image processing mode recognition. The method comprises the steps of: installing a gyroscope and an LED lamp on the target plate, fixing the target plate to the helmet, and placing the left camera and the right camera at the positions as high as the target plate; enabling the DSP to calculate coordinates of the LED lamps in the images collected by the two cameras respectively; according to the conversion relation of the coordinate system, obtaining attitude angle information of the target plate relative to the left camera and the right camera through calculation; solving the azimuth angle and the pitch angle of the initial position, the azimuth angle and the pitch angle of the real-time position and the conversion quantity of the current position relative to the azimuth angle and the pitch angle of the initial position; according to the relative positions of the target plate and the left and right cameras, fusing left and right images to calculate an azimuth angle and a pitch angle; acquiring the azimuth angle speed and the pitch angle speed of the gyroscope in real time; calculating cumulants of the azimuth angle and the pitch angle of the gyroscope according to the attitude angle output moment; and fusing the azimuth angle and the pitch angle calculated according to the image with the azimuth angle and the pitch angle accumulated by the gyroscope.

Description

technical field [0001] The invention relates to the technical field of image processing pattern recognition, in particular to a system in which a servo system follows the movement of the driver's head. Background technique [0002] In recent years, the diversification and intelligence of combat methods on the battlefield have also put forward high requirements for combat assistance systems. When the driver observes the external environment inside the armored vehicle (in a fully enclosed environment), he needs to use the photoelectric system to obtain a normal or even wider driving field of vision; The head posture controls the photoelectric system follow-up system. [0003] In the existing technologies at home and abroad, there are usually three methods: one is to collect and analyze the features of the face, based on the feature point detection of the face, and calculate the motion posture according to the rotation change of the feature points. The system based on this pr...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a multi-sensor fusion-based attitude calculation method to solve the problems of inability to work normally and real-time performance when the head rotation angular velocity is high

Images