Online automobile glass coating type detection device, method and system

Abstract

The invention discloses an online automobile glass coating type detection device, method and system. Three-dimensional data information, that is, thresholds of a width, a height and a cross-sectionalarea, of the automobile windshield glass coating type are preset; a real-time coating type image during the automobile glass coating process is photographed, the position of the coating type is positioned, and three-dimensional image data, that is, the width, the height and the cross-sectional area of the image are extracted; the extracted width, the extracted height and the extracted cross-sectional area of the image are compared with a preset width, a preset height and a preset cross-sectional area, and differences are obtained; whether the differences are in a preset range is judged, and ifyes, automobile windshield glass coating is continued until the coating is completed; or otherwise, the coating type is judged to be unqualified and alarm information is outputted. Environmental light is not required, and the applicability is strong.

Description

Technical field [0001] The invention relates to the field of automobile manufacturing, in particular to an on-line detection device, method and system for automobile glass glue coating. Background technique [0002] With the continuous advancement of automobile manufacturing technology and the continuous optimization of automobile manufacturing processes, automobile glue coating technology has been more and more widely used in major automobile and automobile parts manufacturers. The installation of a car windshield is an important process of a car assembly. The installation quality of the windshield directly affects the airtightness of the vehicle and the safety of the driver, and is an important indicator of car quality. Therefore, an industrial robot system was introduced during the glue installation process of the automobile windshield; in addition, the quality of the glue has a serious impact on the product quality and driving safety of the car, and the manufacturer's intelli...

AI Technical Summary

Problems solved by technology

Two-dimensional vision is based on the contrast (color or grayscale difference) of two-dimensional images to obtain information such as the position of the colloid. position offset

[0004] 1) Only two-dimensional information can be obtained, that is, the colloid width, but the colloid height and cross-sectional area information that are really related to the functional properties of the colloid cannot be obtained;

[0005] 2) In the actual production environment, two-dimensional vision is easily affected by the color of the tested part and ambient light, and its misjudgment rate is very high

[0006] 3) In some special cases, the 2D camera cannot recognize, for example, when the measured part shakes during the glue coating process or the robot path planning is not good, the glue gun may scratch the surface of the part and leave a thin layer of glue marks (But in fact, the amount of glue applied is insufficient), resulting in misjudgment by the 2D vision system and resulting in unqualified parts

[0007] 4) Due to the limitations of the space around the glue gun and the arbitrariness of the glue application direction, the general-purpose single laser line sensor cannot meet the detection requirements, such as the corner arc of the glass, etc., which cannot be detected by a single laser line sensor

Images