Yarn hairiness detection scheme based on machine vision

The hairiness index plays a crucial role in assessing the quality of yarn and serves as an essential indicator of the textile process and yarn processing components. The length, quantity, and distribution of hairiness significantly impact the efficiency and quality of weaving and knitting, while also directly influencing the appearance and market value of the final product. Key evaluation metrics include the number, length, and surface area of hairiness. Manual detection methods are often limited by environmental conditions and high labor intensity, resulting in low efficiency and inconsistent results. To address these challenges, Xian Kede has introduced a machine vision-based approach for detecting hairiness during the production of cop yarn. Their method incorporates an improved median filter combined with the maximum between-class variance technique, effectively reducing image distortion and improving the accuracy of hairiness detection. A digital system has been developed to capture and process images of casing yarn hairiness. The process begins with grayscale transformation and stretching to enhance image contrast. A median filter is then applied to reduce noise and improve image clarity. Finally, the maximum cluster-like variance method is used to generate a clear binary image of the hairiness, enabling accurate measurement of hairiness count, length, and overall quality assessment. Currently, digital image processing is widely used in hairiness detection, with promising results. However, two main issues remain: first, the uneven lighting on the yarn surface can interfere with feature extraction; second, the upper and lower sections of the cop yarn require different preprocessing steps. To overcome these challenges, Xie Kede's automatic appearance detection system uses an enhanced median filter that prevents image distortion and detail loss, while the maximum between-class variance method ensures precise segmentation of the hairiness. The HDGS-I yarn hairiness acquisition system consists of three main components: an industrial camera, LED light source, and a computer. These elements work together to capture high-quality images of the yarn, ensuring clarity and accuracy in hairiness analysis. To maintain image clarity, the camera settings are carefully calibrated. The exposure time is set to 800 μs, and the resolution is set to 800 × 1000 pixels. The camera’s center is aligned with the center of the bobbin to minimize distortion. The distance between the cop and the lens is kept at approximately 20 cm, and the brightness, focal length, and aperture of the light source are adjusted to ensure optimal image quality. Through this system, Xian Kede has developed a digital image acquisition setup for sampling and analyzing yarn hairiness. Grayscale transformation and stretching are used to enhance the contrast between the hairiness and the background. An improved median filter is then applied to remove noise and produce a clean binary image. Finally, the maximum cluster-like variance method is used to extract detailed hairiness information. Experimental results confirm that the system accurately measures the number and length of hairs, offering a reliable solution for modern textile quality control.

10-Port USB/type-c Multi-Port Smart Power Adapter

Multi-Port Adapter,10-port usb multi-port smart power adapter,type-c smart charger

shenzhen ns-idae technology co.,ltd , https://www.best-charger.com