Device and method for acquiring real-time cloth cover image of circular weft knitting machine

Abstract

The invention discloses a device and method for acquiring a real-time cloth cover image of a circular weft knitting machine. The device comprises a machine table, wherein a rotating circular needle cylinder 1 is fixed on the machine table, a cloth roll stand is arranged below the rotating circular needle cylinder 1, the cloth roll stand is connected with the rotating circular needle cylinder 1 through a cloth roll stand linkage bar, a circular ring hanger rod which penetrates through the rotating circular needle cylinder 1 is arranged on the machine table, a circular ring supporting frame that is located below the rotating circular needle cylinder 1 is fixed at the lower end of the circular ring hanger rod, and non-plane to plane supporting rods are arranged on two sides of the circular ring hanger rod; a camera of which the location is inward is fixed at the outer side of the bottom surface of the rotating circular needle cylinder 1, and a stepping cradle head is arranged at the top of the camera; a cylindrical cloth cover is supported by the non-plane to plane supporting rods to form a plane cloth cover, the relative location of the plane cloth cover and the camera remains unchanged to ensure that the camera and the formed plane cloth cover keep the consistent focal length, and thus the acquired image is non-deformed, the texture is clear and consistent, and the brightness is consistent; and furthermore, by adopting the cloth roll stand linkage bar with a 'U'-shaped gap, multiple cameras can obtain the cloth cover images behind the cloth roll stand linkage bar, and the missed detection of cylindrical cloth covers cannot be caused.

Description

technical field [0001] The invention relates to a knitting machine, especially a device and method for acquiring a real-time image of a cloth surface of a circular weft machine. Background technique [0002] During the weaving process of the circular weft machine, since the artificial naked eye is used to check whether there is a defective cloth at intervals on the machine table, it is easy to miss the defective cloth, which leads to the fact that there is a whole roll when the cloth is re-inspected by the cloth inspection machine offline. Cloth is scrapped, which leads to the problem of requiring a lot of manpower; in the prior art, a camera is directly used to acquire images of cylindrical cloth, and the acquired images are deformed, distorted, unclear and consistent in texture, and inconsistent in light and shade. The system processing host first restores the deformed image to the undistorted image through image preprocessing, and uses median filtering, nearest neighbor i...

AI Technical Summary

Problems solved by technology

[0002] During the weaving process of the circular weft machine, since the artificial naked eye is used to check whether there is a defective cloth at intervals on the machine table, it is easy to miss the defective cloth, which leads to the fact that there is a whole roll when the cloth is re-inspected by the cloth inspection machine offline. Cloth is scrapped, which leads to the problem of requiring a lot of manpower; in the prior art, a camera is directly used to acquire images of cylindrical cloth, and the acquired images are deformed, distorted, unclear and consistent in texture, and inconsistent in light and shade. The system processing host first restores the deformed image to the undistorted image through image preprocessing, and uses median filtering, nearest neighbor interpolation algorithm, bilinear difference, high-order interpolation, and restores the image. These image preprocessing processes need It takes a lot of time and increases the hardware cost and the complexity of the circuit, and it is impossible to realize the real-time processing of the image, but the inconsistency of the image light and shade has not always been well resolved; and it is impossible to obtain the covered part of the fabric behind the linkage rod of the cloth roll rack image, so that there is a situation of missed detection on the cloth surface

Images