Machine vision-based cable on-line detection method

Abstract

The invention discloses a machine vision-based cable on-line detection method and belongs to the industrial automation field. The machine vision-based cable on-line detection method is an on-line quality detection method for cable production. The method includes the following steps of: extracting images from video frames, performing image stabilization on the images, performing binaryzation on the images; extracting regions requiring processing through utilizing properties of a histogram; identifying punctiform characters in a character region through adopting a grid method; solving and obtaining lines through utilizing Hough Transfom; and determining defects of a cable according to integral dispersion. The machine vision-based cable on-line detection method has the advantages of high detection speed and high automation degree.

Description

technical field [0001] The invention relates to a cable length recognition detection method and a cable defect detection method in the technical field of machine vision image processing. Background technique [0002] At present, in the cable production line, the counting of the cables is mainly performed by the time sequence of inkjet printing by the inkjet printer. When the count reaches a certain requirement, an interrupt signal is generated to cut off the current cable. However, during the production process of cables, there will be misprints, missing prints, blurred printing, etc., resulting in non-standard production. In addition, the traditional counting method counts according to the printing sequence. When the above phenomenon occurs, the production cannot be stopped immediately, and a batch of products will be scrapped due to part of the printing irregularities. The real-time performance is very poor. In addition, because a certain part of the cable sheath may suf...

AI Technical Summary

Problems solved by technology

However, during the production process of the cable, there will be misprinting, missing printing, blurred printing, etc., resulting in non-standard production.

In addition, the traditional counting method counts according to the printing sequence. When the above phenomenon occurs, the production cannot be stopped immediately, and a batch of products will be scrapped due to part of the printing irregularities. The real-time performance is very poor.

In addition, because a certain part of the cable sheath may suffer losses due to technological reasons, resulting in unqualified products of the same batch, resulting in waste

However, at present, there are no defect detection systems and solutions for cable appearance at home and abroad. The detection of cables usually uses imported systems from abroad to detect characteristics such as insulation, voltage resistance, and resistance.

Images