Infrared multi-point positioning method

Abstract

The invention relates to an infrared multi-point positioning method. The infrared multi-point positioning method comprises the following steps of presetting a window distance threshold value t1, conducting traversal on each viewing angle in the horizontal direction, conducting statics on area information of each viewing angle, and assuming that the number of obtained areas is K1, K2,..., KN respectively; conducting traversal on each viewing angle in the perpendicular direction, conducting statics on area information of each viewing angle, and enabling the number of the obtained areas to be P1, P2,..., PM respectively; calculating a standard score of each point of a candidate point set; conducting statics on a real integral of each point in the candidate point set; judging the true points and the false points. The method has the advantages that the multi-viewing-angle concept is provided; two viewing angles with the most number in the covered area are used as the standard viewing angles, and the standard viewing angles are utilized for calculating candidate touch points; the concept that the minimum distance between the covered area and the points is a standard distance, and other distances are compared with the standard distance is provided; the concept that auxiliary viewing angles are utilized for conducting integration is provided; the four aspects are most important.

Description

technical field [0001] The invention relates to an infrared multi-point positioning method. Background technique [0002] Existing infrared multi-point recognition technology has two kinds of logical judgment method and image processing method, both of which are based on one-to-many hardware scanning method, obtain obstacle boundary according to optical path data; obtain quasi (candidate) touch point set according to boundary , and set the real touch point threshold; traverse each candidate touch point, and determine the credibility of the current point as a real touch point. The reliability is determined as follows. According to the area formed by the boundary of this point, the set of light paths that should theoretically pass through this area is inversely deduced, and the number of actually blocked light path sets is judged in turn. Divide the actual occlusion number by the theoretical occlusion number, if it is greater than the threshold, it is a real point, otherwise ...

AI Technical Summary

Problems solved by technology

Therefore, the image method will not be able to identify the real touch object with the same or close to the grid noise size, even if there is obviously a blocked optical path at this time; in addition, the time difference during high-speed movement will lead to the loss of points after corrosion

The inventors found that the above-mentioned multi-point acquisition scheme has major defects, because the logic method only utilizes the local characteristics of the optical path, and cannot take into account the overall characteristics of the optical path and the mutual influence relationship between real touch points; while the image method only utilizes the global characteristics, Real touch objects that are comparable to noise cannot be recognized, and the image method has a large amount of calculation and memory consumption, is sensitive to delay issues, and is easy to lose real touch points

Images