Detection of residual yarn on spinning bobbins based on salient region detection

Abstract

Residual yarn detector plays an important role in the pipeline of spinning-linked winding systems. This research proposed an image-based method to improve the traditional detectors who have weaknesses such as low precision, low sustainability and yarn-damage-possibility. A detection system was developed to capture and process the bobbin images. The proposed algorithm includes three main steps: bobbin recognition, residual yarn reusability judgment and un-reusable residual yarn detection. With the utilization of the adaptive threshold, profile detection, region-of-interesting extraction and frequency-tuned salient region detection, the bobbins were classified into three classes with a desirable accuracy rate. The proposed method was applied on 21 different bobbin samples and obtained a 100% detection rate, which demonstrated that the method is effective on different samples. To test the robustness of the method, it was tested in eight different light conditions. The result showed that the method is reliable in a wide range of illumination intensity.

Keywords:

• Spinning-linked winding
• yarn bobbin
• residual yarn
• salient region detection

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research was supported by the National Natural Science Foundation of China under Grants 61501209 and 61802152, the National Natural Science Foundation of Jiangsu Province under Grant BK20180602, the Graduate Innovation Project of Jiangsu Province under Grant KYLX16_0789, KYCX18_1819, and National Key R&D Program of China under Grant 2017YFB0309200.