A kind of db type weft yarn detection method

Abstract

The invention discloses a DB type weft yarn detection method. The detection method comprises the following steps: setting a monitoring area and collecting color pictures of textile fabrics; dividing the three-channel image data of the color pictures into three single-channel image data; selecting B-channel image data and convert it from the spatial domain to the frequency domain; remove the noise points in the frequency domain and then convert to the spatial domain; among them, the noise points will be more vivid in the B-channel image data, and the isolated noise points in the spatial domain It is sharper and brighter than the weft edge contour point, so the frequency converted to the frequency domain is larger, and a low-pass filter based on a certain frequency can be selected to suppress high-frequency signals, allow low-frequency signals to pass, and realize noise signals. Then, threshold segmentation is performed on the B channel image data after noise removal, and the boundary points are screened out; the weft yarn is identified according to the continuity of the boundary points.

Description

technical field [0001] The invention relates to the technical field of carbon fiber production, in particular to a DB type weft yarn detection method. Background technique [0002] In the production process of the composite material, the carbon fiber tow and the glass fiber tow need to be woven into a fiber fabric in the warp and weft directions, respectively, and then impregnated with resin and then thermoset. The DB type weft yarn is a layer of weft yarn in the direction of 90 degrees in the production process of the fiber fabric. However, in the actual production process, the number of weft yarns laid out cannot guarantee the actual number of weft yarns and the set number due to equipment reasons. The number of weft yarns needs to be detected in real time during the production process; due to the high degree of automation of the carbon fiber production line, it is difficult to detect the specific number of weft yarns manually during long-term continuous production, and it...

AI Technical Summary

Problems solved by technology

DB class weft yarn is to lay a layer of weft yarn in the direction of 90 degrees during the production process of fiber fabrics, but in the actual production process, the number of weft yarns laid out cannot guarantee the actual number of weft yarns laid due to equipment reasons. To keep the quantity consistent, it is necessary to detect the number of weft yarns in real time during the production process; due to the high degree of automation of the carbon fiber production line, it is difficult to detect the specific number of weft yarns manually during long-term continuous production, and it is easy to appear Missed detection, resulting in unnecessary waste of manpower and material resources

Images