Method for recognising weld spot central position based on computer vision

Abstract

The present invention discloses a computer visual-based method for identifying the central position of the welding point; images are collected by a camera that is arranged above the production line and are inputted into a computer; the computer firstly sets a marking position for the image, so as to calculate the actual deviation of the welding point; then the threshold value of the binaryzation is determined; a binaryzation picture is obtained according to the threshold value, and the area of the welding point is determined according to the binaryzation picture, and finally the center of the welding point is identified; the actual physic coordinate of the welding point center is determined, and the actual physic coordinate is transmitted to the following mechanical device, so as to provide the positioning information. Based on the automatic identification technology, the present invention performs the automatic identification on the central point of the welding point, and has low requirement on the hardware of the computer, thereby reducing the price of the device, and being particularly applicable to the middle and the small enterprises.

Description

technical field [0001] The invention belongs to the field of welding technology, and relates to a method for the center position of a solder joint, in particular to a method for identifying the central position of a solder joint based on computer vision. Background technique [0002] Accurate and efficient identification of solder joints on workpieces is the key to complete automatic welding in large-scale welding automation production lines. [0003] At present, the identification of solder joints on workpieces is mostly done manually. The method of automatic identification is based on digital X-ray real-time imaging, and uses gray scale and shape information to identify the center position of solder joints. This method largely overcomes the limitations of other methods on the test path, and can obtain satisfactory recognition results. However, this automatic recognition method has high requirements for computer hardware and high equipment cost, and is not suitable for sma...

AI Technical Summary

Problems solved by technology

This method largely overcomes the limitations of other methods on the test route, and can obtain satisfactory recognition results. However, this automatic recognition method has high requirements for computer hardware and high equipment cost, and is not suitable for small and medium-sized enterprises.

Images