A Head Pose Recognition Method Based on Data Fusion

Abstract

The invention discloses a head posture recognition method based on data fusion, which first collects the accurate head rotation angle of the X axis and the accurate head Y axis rotation angle during the head rotation process, and then collects the Z axis rotation angle during the head rotation process The Z-axis rotation angle includes the Z-axis rotation angle of the gyroscope and the Z-axis rotation angle of the infrared feature point. The Z-axis rotation angle of the gyroscope and the Z-axis rotation angle of the infrared feature point are used as the input of the data fusion algorithm, and the accurate head is obtained through data fusion. Z-axis rotation angle of the head, and finally output the head posture parameters composed of the accurate head X-axis rotation angle, the accurate head Y-axis rotation angle and the accurate head Z-axis rotation angle; the invention not only has low measurement cost, but also can , can use the infrared scanning measurement results to correct the drift of the gyroscope measurement results, and achieve the purpose of high-precision measurement on the basis of maintaining the original advantages of the two measurement methods.

Description

technical field [0001] The invention relates to the technical field of flight system simulation and space positioning, in particular to a head posture recognition method based on data fusion. Background technique [0002] Most modern fighter jets are equipped with a helmet-mounted aiming system, which allows the pilot to control weapon aiming and radar scanning by turning his head. Before the pilot formally trains on the real machine, he must undergo strict helmet aiming training in the flight simulator. Due to the complex functions and high cost of the real machine helmet aiming system, in the flight simulator, if the pilot wants to turn his head To control functions such as weapon aiming and radar scanning, it is necessary to develop a low-cost helmet aiming system simulation device, in which the core device is the head rotation angle measurement device. [0003] At present, there are two commonly used head rotation angle measurement devices, one is to measure the head ro...

AI Technical Summary

Problems solved by technology

The former can measure the head rotation angle and has a large measurement range, but the data has drift, and the longer the use time is, the greater the drift will be; the latter can also measure the head rotation angle, and the measurement results are highly accurate, but in the case of only one infrared scanner Lower measurable range is small

Images