System and method of online vision detection

Abstract

The invention discloses a system of online vision detection. The system is used for separating good products of core columns on a tension silk machine from poor products, and comprises a human-machine operation interface, a position sensor, an image capturing and detecting device, a main controlling device and a poor product collecting device. The human-machine operation interface is connected with the main controlling device. The image capturing and detecting device is connected with the main controlling device for capturing, analyzing and processing images so as to acquire detecting results. The main controlling device is connected with the position sensor and the poor product collecting device for sending detecting commands according to triggering signals of the position sensor so as to control the actions of the poor product collecting device according to the detecting results. By the aid of the poor product collecting device, the poor products of the core columns are thrown out of a production line, so that separation of good products and poor products is realized. Manual selection is omitted on the production line of tension silk during the whole procedure, full-automatic real-time detection is realized; and real-time performance of poor product detection is improved while undetected rate and detection error on the poor products of the core columns of the tension silk machine are considerably reduced.

Description

Technical field [0001] The invention relates to the field of automatic silk-tensioning machines in the lighting industry, in particular to an online visual inspection system and method for automatically identifying a core column and separating genuine and waste products. Background technique [0002] In the production line of the automatic stretcher in the lighting industry, due to the limitations of the machine itself, raw materials and unexpected unexpected events on site, there will be double core filaments, multiple guide wires, curved guide wires, missing mercury bands, and mercury With various waste products such as deformation and poor electronic powder coating, it is necessary to separate the waste products from the genuine products. However, in the prior art, a method of directly connecting to a low-voltage and manual hand sieve is usually adopted to screen out waste products from genuine products through visual identification. The method of direct access to low voltage...

AI Technical Summary

Problems solved by technology

The method of directly connecting the low voltage can only judge the on-off of the filament, which has great defects; and due to the limited ability of manual identification, the missed detection rate of the waste core column is very high, and it is difficult to accurately detect the waste products on the production line in real time. Screening will seriously affect the qualification rate and material utilization rate of the next process and the entire product line

Images