Tactile-Based Fabric Defect Detection Using Convolutional Neural Network With Attention Mechanism

This paper proposes a fabric structure defect detection method based on the vision-based tactile sensor. The result will be robust by using sensor regardless of dyeing patterns which can influence if some other sensors are used, e.g., vision perception. It also reduces ambient light detection. Therefore, proposed more and universal than conventional visual methods. A robotic arm equipped with was used to automate standardize data collection process construct datasets. In addition, convolutional neural network integrated attention mechanism in channel domain developed detect types. employed frequency filtering remove or weaken normal texture information improve efficiency accuracy. Finally several experiments were conducted demonstrate method’s superiority for detecting structural defects. is evaluated. Experimental results show that feasible efficient meet real-world requirements.