Defects in the textile manufacturing process lead to a great waste of resources and further affect the quality of textile products. Automated quality guarantee of textile fabric materials is one of the most important and demanding computer vision tasks in textile smart manufacturing. This survey presents a thorough overview of algorithms for fabric defect detection. First, this review briefly introduces the importance and inevitability of fabric defect detection towards the era of manufacturing of artificial intelligence. Second, defect detection methods are categorized into traditional algorithms and learning-based algorithms, and traditional algorithms are further categorized into statistical, structural, spectral, and model-based algorithms. The learning-based algorithms are further divided into conventional machine learning algorithms and deep learning algorithms which are very popular recently. A systematic literature review on these methods is present. Thirdly, the deployments of fabric defect detection algorithms are discussed in this study. This paper provides a reference for researchers and engineers on fabric defect detection in textile manufacturing.

/pdf/fabric-defect-detection-in-textile-manufacturing-a-survey-of-25vr9p34dx.pdf

Fabric Defect Detection in Textile Manufacturing: A Survey of the State of the Art

Calibration is required to determine the frequency characteristics of vibration sensors to ensure their measurement accuracy in engineering applications. Thus, a monocular vision-based calibration method for low-frequency vibration sensors is investigated in this study. A sub-pixel edge detection method which based on Gaussian curve fitting is applied to extract the edges of motion sequence images in order to accurately measure the excitation acceleration of the sensors. Because the motion sequence images and the output signal of the sensors cannot be collected synchronously, it is very difficult to align the excitation acceleration signal obtained from the images and the output signal in the time domain. Although the misalignment only has a negligible effect on the magnitude frequency characteristic calibration, it dramatically decreases the calibration accuracy of the phase frequency characteristic, especially with the increasing of the frequency. A time-spatial synchronization technique is proposed to accurately calibrate the phase frequency characteristic by determining the phase of the excitation acceleration signal at a specific spatial position and that of the output signal at the corresponding time. Finally, both the magnitude and phase frequency characteristics are simultaneously calibrated by the investigated method with a flexible and low-cost vision system. The experimental results compared with laser interferometry show that the investigated method accomplishes the high-accuracy magnitude and phase frequency characteristics calibration in a broad low-frequency range. Its calibration accuracy is superior to that of laser interferometry when the frequency is less than 0.3 Hz, and is equivalent at other frequencies.

Monocular Vision-Based Calibration Method for Determining Frequency Characteristics of Low-Frequency Vibration Sensors

Machine learning (ML) and soft computing (SC) techniques have contributed immensely towards improvisation of manufacturing, process and quality control, automation of operations, and decision making in the textile and clothing supply chain. The current study systematically reviews 312 research articles published in Scopus indexed journals in last two decades (2000–2020). Bibliometric analysis revealed a drastic rise in number of studies since 2015. Operations related to manufacturing and quality control of yarns and fabrics are found to be benefitted most from ML techniques, as deduced from keyword and thematic content analyses. On the other hand, with respect to ML and SC techniques, artificial neural network, genetic algorithm and fuzzy logic have preponderance over the others. Nevertheless, over the years, many novel ML and SC techniques have emerged in accomplishing diverse objectives of the textile and clothing supply chain. A detailed analysis of their evolution has been documented. Unfortunately, the extent of knowledge sharing and collaboration within the scientific community working in this domain has been very low. This review also unearths lack of integrated focus on complete value chain of textiles and clothing. Fibre development, fuzzy inference and control in textile machines, predictive analytics for clothing engineering, and fashion forecasting have been identified as areas for future research. 

Machine learning and soft computing applications in textile and clothing supply chain: Bibliometric and network analyses to delineate future research agenda

A monocular vision-based decoupling measurement method for plane motion orbits

Calibration is required to maximize the sensitivity and measurement accuracy of vibration sensors. In this study, a low-frequency vibration calibration method is proposed that is based on the concept of monocular vision. In this method, we employ a high-accuracy edge extraction method to extract the edges of sequential images so as to obtain the high calibration accuracy. However, the proposed method must rely on a long-stroke shaker to provide vibration excitation to the sensor, and the bending in the guideway caused by the mechanical processing reduces the calibration accuracy, especially at very low frequencies. The proposed setting compensates for the bending using an additional monocular vision technique to significantly improve the calibration accuracy. To validate the calibration accuracy of the proposed method, a comparison was conducted between results obtained via the laser interferometry, the Earth's gravitation method, and the proposed method when applied to calibrate the sensitivity of a tri-axial acceleration sensor at frequencies between 0.04 and 8 Hz. The results of the comparison showed the proposed method calibrated the sensor sensitivity with high accuracy and was able to accurately account for the bending when the frequency was lower than 0.3 Hz. In contrast, the calibration accuracy of the laser interferometry decreased because of the bending.

Monocular vision-based low-frequency vibration calibration method with correction of the guideway bending in a long-stroke shaker.

Heterodyne interferometers have been widely used for primary vibration calibration in recent years. Primary vibration calibration performance depends on the precision and real-time nature of vibration measurements, factors determined by the acquisition and demodulation of the heterodyne interferometer signal. This signal is commonly collected using the Nyquist sampling method, requiring devices with high sampling rates and large memories, or a sampling method using a mixer and low-pass filter-analog devices which may create time delays. This study proposes a novel bandpass sampling method that reduces sampling rate and storage capacity without generating time delays. To improve vibration measurement precision, an optimal sampling rate is designed to collect the heterodyne interferometer signal, and the collected signal is demodulated using the phase-unwrapping sine approximation method. The proposed method is compared with existing methods through simulated and experimental data. Experimental results show that the proposed method avoids time delays and high sampling rates, while providing high-precision vibration measurements.

Bandpass-sampling-based heterodyne interferometer signal acquisition for vibration measurements in primary vibration calibration.

Fabric defect detection is very important in the textile quality process. Current deep learning algorithms are not effective in detecting tiny and extreme aspect ratio fabric defects. In this paper, we proposed a strong detection method, Priori Anchor Convolutional Neural Network (PRAN-Net), for fabric defect detection to improve the detection and location accuracy of fabric defects and decrease the inspection time. First, we used Feature Pyramid Network (FPN) by selected multi-scale feature maps to reserve more detailed information of tiny defects. Secondly, we proposed a trick to generate sparse priori anchors based on fabric defects ground truth boxes instead of fixed anchors to locate extreme defects more accurately and efficiently. Finally, a classification network is used to classify and refine the position of the fabric defects. The method was validated on two self-made fabric datasets. Experimental results indicate that our method significantly improved the accuracy and efficiency of detecting fabric defects and is more suitable to the automatic fabric defect detection.

Ying Wang

Papers

Monocular vision-based low-frequency vibration calibration method with correction of the guideway bending in a long-stroke shaker.

Monocular Vision-Based Calibration Method for Determining Frequency Characteristics of Low-Frequency Vibration Sensors

Bandpass-sampling-based heterodyne interferometer signal acquisition for vibration measurements in primary vibration calibration.

A monocular vision-based decoupling measurement method for plane motion orbits

Automatic Fabric Defect Detection Method Using PRAN-Net