Dimensional-histogram-statistic-based touch pen color recognition method

Abstract

The invention discloses a dimensional-histogram-statistic-based touch pen color recognition method, which comprises: setting a searching template to search for a black border range of a touch device in an image taken by a camera; using the region, in which the searching template is positioned and in which the sum of the red-green-blue (RGB) values of all pixels is maximum, as the region in which the pen head of the touch pen is positioned, and determining a sampling region by using the central point in the region as a center of a circle and by using a set pixel point as a radium; dividing the sampling region by setting a saturation threshold; enhancing the colors of the divided images; and determining the color of the touch pen by dimensional histogram statistic technology. In the invention, the adoption of color enhancement technology improves the anti-noise capacity and the adoption of the dimensional histogram statistic technology effectively avoids the influences of white-pixel-oriented display of the body of the pen in the image due to the light radiation of the body of the pen, so that the touch pen color recognition accuracy is high and the region wherein the pen head of the touch pen is positioned can be determined accurately and effectively; and the method also has the advantages of simplicity of operation, high processing speed, and the like.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a touch pen color recognition method based on histogram statistics. Background technique [0002] At present, the color recognition of the touch pen mainly has the following methods: (1) through magnetic pen recognition, that is, by placing a special magnetic pen on the touch panel to realize pen color recognition, but this method has a high manufacturing cost; (2) Through infrared induction recognition, the touch pen is placed in a special tray, and each tray corresponds to a color. When the infrared induction tray detects that the touch pen is removed from the tray, the color of the touch pen is confirmed, but this method The disadvantage is that it is easy to make mistakes because the touch pen is placed on the wrong tray, and it is necessary to realize that the number of recognized colors is proportional to the number of trays, which is not conducive to the increase of touch c...

AI Technical Summary

Problems solved by technology

[0002] At present, the color recognition of the touch pen mainly has the following methods: (1) through magnetic pen recognition, that is, by placing a special magnetic pen on the touch panel to realize pen color recognition, but this method has a high manufacturing cost; (2) Through infrared induction recognition, the touch pen is placed in a special tray, and each tray corresponds to a color. When the infrared induction tray detects that the touch pen is removed from the tray, the color of the touch pen is confirmed, but this method The disadvantage is that it is easy to make mistakes because the touch pen is placed on the wrong tray, and it is necessary to realize that the number of recognized colors is proportional to the number of trays, which is not conducive to the increase of touch colors; (3) the touch pen is photographed by the camera. The method extracts the color of the touch pen and realizes pen color recognition. This method is popular because it can recognize any color and has the advantages of low cost. The image captured without a touch pen and the captured image with a touch pen are subtracted, and the position of the maximum difference is used as the position of the touch pen, and then the color of the image in a certain area near the position is extracted as the color of the touch pen

This method has two disadvantages: on the one hand, using the difference operation, it is easy to cause the obtained position to be only the place where the image noise is superimposed, rather than the real position of the touch pen; on the other hand, the average value of the image color in a certain area is taken as the The accuracy of the pen color is not high, especially when the pen body is illuminated by light, a large part of the pen body will appear off-white in the image. If the average value of the image color in a certain area is simply taken as the pen color, it will There is a big error with the actual pen color

Images