top of page

Increased Efficiency with Smart Fabric Control Systems

Writer's picture: Ayşegül YıldızAyşegül Yıldız

Akıllı Kumaş Kontrol Sistemleri

Contributions of Industrial Artificial Intelligence to Textile Production: Automatic Fabric Defect Detection and New Approaches

Artificial Intelligence (AI) has brought significant transformation to the industrial sector, exceeding societal expectations by a wide margin. Numerous engineers and researchers are continuously working to accelerate the development of industrial intelligence. Industrial AI is a multidisciplinary field that integrates data science, mechanical engineering, communication, information security, and network systems. The primary objective of this technology is to address the challenges faced by industries while enhancing the safety and efficiency of production processes. The textile production sector, as part of this transformation, must adopt innovative production methods within the scope of Industry 4.0.


Textile production consists of large-scale and complex processes. The term "textile fabric defect" refers to flaws on the surface of fabrics, which typically arise from machine malfunctions or process issues. Fabric defects reduce the quality of the final product and result in resource wastage. Since each stage of the production process affects the subsequent one, early detection of defects is critical to preventing losses. Consequently, systems that automate defect detection are pivotal for textile manufacturers to improve product quality and control costs.


Challenges of Manual Inspection and the Advantages of Automation

Traditionally, fabric defect detection has been performed through manual visual inspection. However, this method is both costly and limited in efficiency. The main challenges include:


  1. Fabric inspectors find it difficult to maintain focus on the production process for extended periods, which reduces efficiency due to lack of attention.

  2. Inspectors require training, which demands both time and financial resources.

  3. Small defects often go unnoticed, with only prominent flaws being identified.

  4. Locating defects is a labor-intensive process.


Research indicates that the accuracy rate of manual inspection ranges between 60% and 75%. This increases waste and raises product costs. In this context, the use of automatic defect detection mechanisms in fabric production processes emerges as a critical solution for reducing labor costs and improving quality.



The Role of Automatic Fabric Defect Detection Systems

Automatic fabric defect detection systems revolutionize quality control processes by integrating digital image processing and artificial intelligence techniques. These systems enable rapid and accurate defect detection by being seamlessly integrated into real-time production processes. In particular, digital image processing and computer vision (CV) technologies play a crucial role in these processes.


A Next-Generation Intelligent Fabric Defect Detection Method

This study introduces a hybrid artificial intelligence and deep learning-based fabric defect detection method to ensure sustainability and enhance quality in fabric production. The proposed method improves the quality of fabric images, generating optimized visuals for analysis. Subsequently, a feature extraction process is applied to identify defects, and this information is classified using artificial neural networks. The system automatically identifies the type and location of defects, making production processes more efficient.

The proposed method has been tested on an open-access dataset, demonstrating high accuracy rates in rapidly detecting fabric defects. This once again highlights the growing importance of digitization and automation in the textile industry.

2 views
whatsapp-icon.png
bottom of page