Deep Learning-Based Smart Colored Fabric Defect Detection System
01 Jan 2020-pp 212-219
TL;DR: The working and reliability of the fabric defect detection system presented is evaluated through vigorous experiments of real fabric samples with different defects, using the concept of deep learning for the detection of colored fabric defects.
Abstract: Due to the huge increase in customers for different fabrics in this generation, the texture of the fabrics becomes an important issue thus bringing the requirement for correct and perfect detection of the fabric defects. In the existing semiautomated systems, a quality inspector takes 5 m/min with a defect and 15 m/min without defect to identify and rectify the defects with the resolution of 1 mm/pixel. This process results in the loss of the factory’s overall throughput and efficiency. While manufacturing fabrics, there may be various defects like hole, missing yarn, broken yarn, stain, etc. These defects incur huge losses to the textile industry as they cause customer dissatisfaction. In order to reduce such losses, detection of defects beforehand is very important. Our project uses the concept of deep learning for the detection of colored fabric defects. The working and reliability of the fabric defect detection system presented is evaluated through vigorous experiments of real fabric samples with different defects.
Citations
More filters
TL;DR: In this paper, the authors presented a thorough overview of algorithms for fabric defect detection in textile smart manufacturing, including traditional algorithms and learning-based algorithms, and traditional algorithms are further categorized into statistical, structural, spectral, and model-based methods.
Abstract: 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.
36 citations
27 Jan 2021
TL;DR: In this paper, a convolutional transformer-based NLG-based chatbot was used to identify people from their unique style of chatting in natural / casual conversations, which is able to capture these traits in conversation for short responses.
Abstract: In natural / casual conversations, people tend to have their own unique style of speaking/ chatting; certain phrases only they use, style of typing, customized grammar, nicknames etc. These ‘imperfections’ make the dialogue more human; we are able to identify people from their unique style of chatting. Conversational chatbots trained on clean text won't be able to capture these nuances. We propose a method to capture these quirks in the form of a chatbot with a novel NLG (Natural Language Generation) architecture; In the form of a convolutional transformer. WhatsApp chats between two users were cleaned and exported for training the model. The dataset had Unicode characters for foreign lexicon, emojis, and personalised grammar / texting patterns. The model is able to capture these traits in conversation for short responses.
4 citations
TL;DR: In this paper , a custom Convolution Neural Network (CNN) was used for skin cancer classification on a Human-Against-Machine (HAM10000) database which is publicly accessible through the Kaggle website.
Abstract: Melanin skin lesions are most commonly spotted as small patches on the skin. It is nothing but overgrowth caused by melanocyte cells. Skin melanoma is caused due to the abnormal surge of melanocytes. The number of patients suffering from skin cancer is observably rising globally. Timely and precise identification of skin cancer is crucial for lowering mortality rates. An expert dermatologist is required to handle the cases of skin cancer using dermoscopy images. Improper diagnosis can cause fatality to the patient if it is not detected accurately. Some of the classes come under the category of benign while the rest are malignant, causing severe issues if not diagnosed at an early stage. To overcome these issues, Computer-Aided Design (CAD) systems are proposed which help to reduce the burden on the dermatologist by giving them accurate and precise diagnosis of skin images. There are several deep learning techniques that are implemented for cancer classification. In this experimental study, we have implemented a custom Convolution Neural Network (CNN) on a Human-against-Machine (HAM10000) database which is publicly accessible through the Kaggle website. The designed CNN model classifies the seven different classes present in HAM10000 database. The proposed experimental model achieves an accuracy metric of 98.77%, 98.36%, and 98.89% for protocol-I, protocol-II, and protocol-III, respectively, for skin cancer classification. Results of our proposed models are also assimilated with several different models in the literature and were found to be superior than most of them. To enhance the performance metrics, the database is initially pre-processed using an Enhanced Super Resolution Generative Adversarial Network (ESRGAN) which gives a better image resolution for images of smaller size.
1 citations
23 Mar 2023
TL;DR: In this article , an Improved Dragonfly optimization with Deep Learning based Fabric Defect Classification (IDFODL-FDC) technique was proposed to classify the fabric images as defective or not.
Abstract: The process of recognizing abnormalities or defects in fabric is termed Fabric defect detection. Fabric defects may arise because of several factors like wear and tear, manufacturing errors, and damage during transportation. Finding such defects is vital to guarantee the durability and quality of the finished product. Among several image classifier tasks, Deep learning (DL) is considered a successful one and has been implemented in fabric defect detection. Therefore, this study develops an Improved Dragonfly optimization with Deep Learning based Fabric Defect Classification (IDFODL-FDC) technique. The proposed IDFODL-FDC technique categorizes the fabric images as defective or not. To do this, the IDFODLFDC technique employs feature extractor based on MobileNet model. In addition, the hyperparameter tuning of the MobileNet model performed via the IDFO technique. Finally, extreme gradient boosting (XGBoost) method is used for fabric defect classification. A series of experiments have been conducted to reveal the enhanced performance of the IDFODL-FDC method. The simulation values stated that the IDFODL-FDC method for fabric defect detection demonstrated good performance and has the potential to increase the efficiency and accuracy of quality control processes in the textile industry.
26 Aug 2022
TL;DR: In this article , the authors developed a model of a drowsiness detection system that can track whether the driver's eyes are open or closed in real-time, based on eye localization, which entails scanning the captured image to determine the eyes' location using the facial landmarks algorithm with the help of Dlib and OpenCV.
Abstract: Driver exhaustion can be an essential factor in an excessive number of vehicle accidents. Detection technology or avoiding drowsiness at the wheel may be a significant challenge in collision avoidance systems. This research aims to develop a model of a drowsiness detection system that can track whether the driver's eyes are open or closed in real-time. It is thought that early identification of signs of driver exhaustion, such as watching the eyes, is often enough to avoid a road accident. The detection of fatigue necessitates a series of eye photographs to observe the eye movements and blink patterns. The proposed system is based on eye localization, which entails scanning the captured image to determine the eyes' location using the facial landmarks algorithm with the help of Dlib and OpenCV. It determines whether they are open or closed to detect drowsiness, adding a feature for the driver assistant system to avoid collisions.
References
More filters
01 Jan 1998
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Abstract: Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.
42,067 citations
TL;DR: The proposed method was successfully applied to Brodatz mosaic image segmentation and fabric defect detection and can be expanded to an unsupervised texture segmentation using a Kullback-Leibler divergence between two Gaussian mixtures.
200 citations
TL;DR: A comprehensive literature review of fabric defect detection methods, categorized into seven classes as structural, statistical, spectral, model-based, learning, hybrid and comparison studies, finds weaknesses of each approach.
150 citations
TL;DR: A deep-learning algorithm was developed for an on-loom fabric defect inspection system by combining the techniques of image pre-processing, fabric motif determination, candidate defect map generation, and convolutional neural networks (CNNs).
Abstract: Loom malfunctions are the main cause of faulty fabric production. A fabric inspection system is a specialized computer vision system used to detect fabric defects for quality assurance. In this paper, a deep-learning algorithm was developed for an on-loom fabric defect inspection system by combining the techniques of image pre-processing, fabric motif determination, candidate defect map generation, and convolutional neural networks (CNNs). A novel pairwise-potential activation layer was introduced to a CNN, leading to high accuracy of defect segmentation on fabrics with intricate features and imbalanced dataset. The average precision and recall of detecting defects in the existing images reached, respectively, over 90% and 80% at the pixel level and the accuracy on counting the number of defects from a publicly available dataset exceeded 98%.
88 citations