The feature extraction of trace is also the precondition and one of the important criteria for target tracking. The algorithm with large amount of feature extraction and strong accuracy can avoid tracking errors and have better quality when tracking. In order to improve the comparison ability of the experiment, the contact trace tracking method of spiral bevel gear and the contact analysis method of spur bevel gear pair based on installation and adjustment will be compared and analyzed. The experimental results are shown in the figure.
It can be seen from Fig. 8 that the method in this paper has the best extraction effect on trace features, the highest curve distribution range and strong stability; The accuracy of feature extraction of the other two methods is low, showing a relatively unstable trend as a whole, and with the continuous increase of data volume, the extraction accuracy continues to decline, which indicates that the adaptability of the algorithm is not strong, and the experience is insufficient in the face of a relatively large amount of data.
This is mainly because they do not analyze the motion of spiral bevel gears in detail, resulting in the algorithm’s weak sensitivity to the operation of spiral bevel gears, lack of capture characteristics when trace features appear, and poor natural extraction effect.