An Uncertainty-Aware Deep Learning Framework for Defect Detection in Casting Products

Purpose: The aim of the article is to develop an algorithm for classifying cracks in the analyzed images using modern methods of deep machine learning and transfer learning based on pretrained convolutional neural network - Inception-ResNet-v2. Design/Methodology/Approach: Transfer learning based on the pretrained convolutional neural network was used to categorize the images. The fully conected layer of the Inception-ResNet-v2 network has been modified. The last layer was trained using a two-class (binary) linear SVM (Support Vector Machine). In total, 20,000 training cases (images) were used to train the fully connected layer within transfer learning process. The research analyzed the possibility of using the deep neural networks for quick and fully automatic identification of cracks / defects on the surface of analyzed parts. Findings: The results indicate that pretrained convolutional neural network using SVM to train a fully connected layer is a very effective solution for visual crack / fault detection. In the analyzed model, a positive classification was obtained at the level of 99.89%. Practical Implications: The model presented in the article can be used in quality control carried out by process monitoring. An effective model for identifying defective parts can be used in both logistics and production processes. Originality/Value: A novelty is the use of a freely available, deep neural network trained to classify 1000 categories of various images for binary categorization of faults (cracks). The algorithm was adjusted by replacing the primary, 1000-output fully connected layer in the Inception-ResNet-v2 network with a binary layer (2 categories). The fully connected layer has been trained using the classification version of the popular SVM learner, but thanks to the combination of this layer with the sophisticated fearure extraction ability of the pre-trained Inception-ResNet-v2 deep network, the resulting predictive model enables the classification of defects with a very high level of accuracy.

Emotion recognition plays a pivotal role in advancing Human-Computer Interaction (HCI) by enabling systems to understand and respond to human emotional states. Among various modalities, Speech Emotion Recognition (SER) stands out as a non-invasive, cost-effective, and temporally efficient approach for detecting emotions. This study leverages the RAVDESS dataset, which provides high-quality audio recordings for emotion classification. The proposed methodology involves preprocessing audio signals to remove noise, extracting temporal and spectral features in both time and frequency domains, and implementing machine learning models for multi-class emotion classification. The study evaluates and compares the performance of several classification models, including Random Forest (RF), Multilayer Perceptron (MLP), Support Vector Machine (SVM), Convolutional Neural Networks (CNN), and Decision Trees (DT). Experimental results demonstrate promising accuracy levels, highlighting the potential of these machine learning techniques in SER. This research contributes to the integration of SER within Brain-Computer Interfaces (BCI) and other emotion-aware applications, paving the way for enhanced interactive systems. Keywords – Speech Emotion Recognition -Machine Learning - Random Forest -Multilayer Perceptron - Support Vector Machine - Convolutional Neural Networks - Decision Tree - RAVDESS Dataset

Automatic surface defect detection of mobile phone in large scale needs to process high resolution images and handle various defects while achieving high accuracy rate. This study proposes a defect detection method based on convolution neural network (CNN). Firstly, the original surface image of mobile phone is obtained using industrial linear array camera. Secondly, the obtained images are automatically segmented into specified sizes by the proposed preprocessing step. Moreover, we design the CNN on basis of GoogLeNet network, which greatly reduces the number of parameters without compromising prediction rate. At last the designed CNN are trained and tested. The trained CNN can be combined with a sliding window technique to detect any ROI with size larger than 256×256 resolutions in the original images. The experimental results show that the defect detection rate of the designed CNN can achieve as high as 99.5%.

The quality monitoring of fiber-optic coil (FOC) in winding systems is usually done manually. Aiming at the problem of inefficient and low accuracy of manual detection, this article is dedicated to researching a defect detection framework based on machine vision, which provides a reliable method for automatic defect detection of FOC. For this purpose, a defect detection scheme that integrates wavelet transform and nonlocal means filtering is proposed to accurately locate the defect region. Then, based on the features constructed by wavelet coefficients, a support vector machine (SVM) is used as the classifier. Additionally, a self-adaptive genetic algorithm is proposed to optimize the parameters of the SVM to form the final classifier. Through experiments on the data set obtained by our designed imaging system, the results show that our method has good defect detection performance and high classification accuracy, which provides an optimal solution for the automatic detection of FOC.

Here we investigate the application of convolutional neural networks (CNNs) to predict the survival time of patients with Glioblastoma Multiforme (GBM) brain tumor. Our dataset consists of T1-weighted high-resolution MRI images of just 68 GBM patients. We compare two analytic methods for predicting survival time. The first consists of training a small convolutional neural network (CNN) and the second uses extracted deep features from a pre-trained CNN. Our method is completely automated, except for tumor region segmentation. In addition, we utilize a snapshot ensemble approach to boost test accuracy when dealing with limited availability of medical images for CNN training purposes. Our approach achieves an accuracy of 72.06% using a trained small network and 66.18% using a pre-trained deep CNN. Our results compare favorably with the accuracy of 54.41% using histogram of oriented gradients (HOG) features and a non-neural network classifier.

Automatic defect detection is of great significance to the production process of modern industries, which will affect the product appearance and quality. And a great number of economic loss may be caused due to the defects in the industrial production process. Traditional manual defect inspection method is labor-intensive and time-consuming with subjective factors, and the detection performance is random and uncertain. Machine learning has shown a good detection ability on small-scale samples, but the detect detection task against with poor contrast, weak texture, etc., will affect the effective feature representation. To address the detection task of steel surface defects, combined with the strong context extraction ability of deep learning, an accurate deep defect classification network is proposed in the paper to provide an end-to-end detection scheme. Fused with the residual network (Resnet50) and spatial attention block, a residual attention network is proposed for effective feature representation, which could make the classification network better focus on the defect areas. Meanwhile, due to the scale information change among different defects, a multi-scale context fusion (MCF) block is proposed for effective multi-scale feature extraction, which is conducive for multi-scale object detection. Experimental results on public defect data set show that the proposed defect detection network could acquire a superior classification performance compared with some typical classification networks.

Graph convolutional network – Long short term memory neural network- multi layer perceptron- Gaussian progress regression model: A new deep learning model for predicting ozone concertation

Abstract. In recent years, there has been a significant improvement in the detection, identification and classification of objects and images using Convolutional Neural Networks. To study the potential of the Convolutional Neural Network, in this paper three approaches are investigated to train classifiers based on Convolutional Neural Networks. These approaches allow Convolutional Neural Networks to be trained on datasets containing only a few hundred training samples, which results in a successful classification. Two of these approaches are based on the concept of transfer learning. In the first approach features, created by a pretrained Convolutional Neural Network, are used for a classification using a support vector machine. In the second approach a pretrained Convolutional Neural Network gets fine-tuned on a different data set. The third approach includes the design and training for flat Convolutional Neural Networks from the scratch. The evaluation of the proposed approaches is based on a data set provided by the IEEE Geoscience and Remote Sensing Society (GRSS) which contains RGB and LiDAR data of an urban area. In this work it is shown that these Convolutional Neural Networks lead to classification results with high accuracy both on RGB and LiDAR data. Features which are derived by RGB data transferred into LiDAR data by transfer learning lead to better results in classification in contrast to RGB data. Using a neural network which contains fewer layers than common neural networks leads to the best classification results. In this framework, it can furthermore be shown that the practical application of LiDAR images results in a better data basis for classification of vehicles than the use of RGB images.

Nondestructive thermography is a high-speed, low-cost, and safe solution for subsurface defects detection of carbon fiber reinforced polymer (CFRP) materials, providing essential quality control in aerospace, automobile, and sports industries. In this article, we build a reflective lock-in thermography system and construct a dataset that contains real-captured thermal image sequences of CFRP samples with various simulated internal defects under different excitation frequencies. Then, we present a novel 3-D convolutional neural network (CNN) model incorporating a combination of spatial and temporal convolutional filters and batch-size independent group normalization (GN) as a unified framework to process thermal image sequences captured by lock-in thermography for simultaneous subsurface defect detection and depth estimation. Finally, we define a multitask loss function to perform end-to-end training of both defect detection and depth estimation tasks based on the real-captured infrared sequences. Comparative experiments are carried out on CFRP specimens with artificial defects of various sizes/shapes and at different depths. Qualitative and quantitative results illustrate that our 3-D CNN model is capable of predicting accurate locations and depths of subsurface defects and performs favorably against the hand-crafted and CNN-based methods in lock-in thermography for individual defect detection and depth estimation tasks. The captured dataset and the source codes will be made publicly available.

Scene classification plays a crucial role in the travel and tourism industry in promoting travel. In this paper, we propose a fusion approach based on representation learning to perform scene classification using a basic convolutional neural network (CNN) and multiple pre-trained CNNs as feature extractors. We have used the Oliva and Torralba Scene (OT-Scene) dataset consisting of scene images of eight different categories. In this proposed approach we extract image representations from deep learning models such as basic CNN and pre-trained CNN, namely VGG19, MobileNetV2, and RestNet50. We then fuse these representations from several models into different combinations to achieve a more generalized representation or feature vectors. The fused feature vectors and the individual feature vectors are then fed into the artificial neural network (ANN), support vector machine (SVM), and random forest (RF) classifiers separately to compare the results. The results showed the highest scene classification accuracy of 97.4% from specific fused combinations.

ObjectiveIn this paper, we propose to evaluate the use of pre-trained convolutional neural networks (CNNs) as a features extractor followed by the Principal Component Analysis (PCA) to find the best discriminant features to perform classification using support vector machine (SVM) algorithm for neonatal sleep and wake states using Fluke® facial video frames. Using pre-trained CNNs as a feature extractor would hugely reduce the effort of collecting new neonatal data for training a neural network which could be computationally expensive. The features are extracted after fully connected layers (FCL’s), where we compare several pre-trained CNNs, e.g., VGG16, VGG19, InceptionV3, GoogLeNet, ResNet, and AlexNet.ResultsFrom around 2-h Fluke® video recording of seven neonates, we achieved a modest classification performance with an accuracy, sensitivity, and specificity of 65.3%, 69.8%, 61.0%, respectively with AlexNet using Fluke® (RGB) video frames. This indicates that using a pre-trained model as a feature extractor could not fully suffice for highly reliable sleep and wake classification in neonates. Therefore, in future work a dedicated neural network trained on neonatal data or a transfer learning approach is required.

In order to realize automatic surface defect detection for large aperture precision spherical optical elements, an automatic surface defect detection instrument has been designed. The instrument consists of a microscopic imaging system, illumination system, motion scanning system, and a software algorithm system. Firstly, a multi-angle channel illumination source and a coaxial illumination source were designed. Bright and dark field images of surface defects were captured by cooperating with an automatic zoom microscope. Then, algorithms for scanning trajectory planning, image stitching, and intelligent defect recognition were designed to achieve full-aperture surface image acquisition and defect quantification detection. The automated defect detection process of the instrument is summarized and introduced. Finally, the experimental platform was constructed, which can work well for the optical elements with a maximum diameter of 400 mm and a relative aperture R/D value of 1. It takes about 15 min to detect an optical element with a diameter of 200 mm in dark-field imaging mode. As a result, the minimum line width of scratch detectable is 2 μm and the minimum diameter of pitting detectable is 4 μm. Clearly, the instrument can realize the automatic detection of surface defects of spherical optical elements, and has the advantages of a high efficiency, stability, reliability, quantification, and data traceability.

Introduction: This paper presents a novel approach for classifying different types of steel rods using a combination of pre-trained convolutional neural networks (CNNs) and fuzzy logic systems. We utilize VGG16, ResNet50, and InceptionV3 models to extract features from grayscale images of steel rods, which are then classified using Support Vector Machines (SVM). Furthermore, we incorporate a fuzzy logic system to improve the final classification accuracy by combining the individual accuracies of each CNN model. Experimental results demonstrate the effectiveness of our approach in achieving high classification accuracy compared to traditional methods. Objectives: This research presents a novel multi-model approach for steel rod classification utilizing three distinct fuzzy logic systems: Fuzzy Recurrent Neural Network (RNN), Fuzzy Convolutional Neural Network (CNN), and Fuzzy Neural Network. The proposed system addresses the critical challenge of accurate defect detection and classification in steel manufacturing processes. By combining these complementary approaches, we achieve enhanced feature extraction capabilities and improved classification accuracy. Experimental results demonstrate that our multi-model system achieves a classification accuracy of 97.8%, surpassing single-model approaches by an average of 8.2%. The system successfully identifies surface defects, structural anomalies, and dimensional variations with high precision, making it suitable for real-time industrial applications. Methods: This paper introduces a novel approach for the feature extraction and classification of steel rods by integrating multiple models and utilizing fuzzy logic systems. The methodology leverages pre-trained convolutional neural networks (CNNs) for extracting comprehensive feature representations from steel rod images. These extracted features are then combined and integrated using fuzzy logic to enhance classification accuracy and reliability. By addressing the limitations of traditional manual inspection methods and single-model approaches, this integrated multi-model system offers a robust solution for real-time industrial implementation. The proposed system aims to improve the efficiency and effectiveness of quality control processes in steel rod manufacturing, ensuring consistent and accurate classification even under varying conditions and complexities. This innovative approach paves the way for more advanced and adaptable automated inspection systems in the industry. Results: The paper aims to develop an integrated multi-model system that combines three distinct fuzzy logic approaches to enhance feature extraction capabilities through complementary methodologies. By improving the classification accuracy and reliability, the system addresses the limitations of existing methods. The goal is to create a robust and adaptable system that is suitable for real-time industrial implementation for steel rod analysis and classification. This comprehensive approach leverages the strengths of multiple models and fuzzy logic techniques, ensuring accurate and efficient classification processes, even under complex and varying conditions. The proposed multi model system is designed to meet the stringent quality control requirements of industrial applications, ultimately improving productivity and product quality through reliable automated inspection and classification for steel rod with accuracy of 97.80%, precision of 97.30% and recall 97.50%. Conclusions: This paper presents a novel approach for the classification of steel rods using a combination of pre-trained CNN models and fuzzy logic systems. The experimental results demonstrate the effectiveness of our method in achieving high classification accuracy. This research demonstrates the effectiveness of a multi-model fuzzy logic approach for steel rod classification. The system achieves significant improvements over single-model approaches while maintaining real-time processing capabilities. Our approach provides a foundation for further research and development in automated industrial image classification.

Automated ultrasonic testing (AUT) is a nondestructive testing (NDT) method widely employed in industries that hold substantial economic importance. To ensure accurate inspections of exclusive AUT data, expert operators invest considerable effort and time. While artificial intelligence (AI)-assisted tools, utilizing deep learning models trained on extensive in-laboratory B-scan images, whether they are augmented or synthetically generated, have demonstrated promising performance for automated ultrasonic interpretation, ongoing efforts are needed to enhance their accuracy and applicability. This is possible through the evaluation of their performance with experimental ultrasonic data. In this study, we introduced a real-world ultrasonic B-scan image dataset generated from proprietary recorded AUT data during industrial automated girth weld inspection in oil and gas pipelines. The goal of inspection in our dataset was detecting a common type of defect called lack of fusion (LOF). We experimentally evaluated deep learning models for automatic weld defect detection using this dataset. Our assessment covers the baseline performance of state-of-the-art (SOTA) models, including transformer-based models (DETR and Deformable DETR) and YOLOv8. Their flaw detection performance in ultrasonic B-scan images has not been reported before. The results show that, without heavy augmentations or architecture customization, YOLOv8 outperforms the other models with an F1 score of 0.814 on our test set.

In this study, we used two machine learning algorithms, namely, linear support vector machine (SVM) and convolutional neural network (CNN), to classify the BCI (Brain Computer interface) competition IV-2a 2-class MI (motor imagery) data set which consists of EEG data from 9 subjects. For each subject, 5 sessions of signals from three electrodes (C3, Cz, and C4) were recorded with sampling rate 250Hz. The training data, which consisted of the first 3 sessions, included 400 trials. The evaluation data, which consisted of the last 2 sessions, included 320 trials. Each trial started with gazing at fix cross on screen for 3 seconds followed by a one-second visual cue pointing either to the left or right to instruct the subject for left or right motor imagery over a period of 4 seconds, and then followed by a short break of at least 1.5 seconds. Features were extracted from the 0.5 to 2.5 second signals after the cue for each trial from C3 and C4. Each EEG trial was band pass filtered into different frequency bands, namely, delta (0.5-3Hz), theta (4-8Hz), alpha (8-12Hz), beta bands (13-30Hz), gamma bands (31-60Hz). Those filtered signals were then used as the input data for training the linear SVM. In addition, we generated a 2 by 500 matrix by down sampling the training data from each trial. There are 5760 such matrices in total generated from all subjects and serve as the input data for training CNN and the trained model was evaluated by another 340 matrices from each subject. Our CNN architecture consisted of 2 convolution layer and 2 fully connect layers, and there was a batch normalization layer before the activated layer and a dropout layer with a probability of 50% after the activated layer. The classification accuracies evaluated by averaged kappa values obtained from linear SVM and CNN are 0.5 and 0.621, respectively, suggesting the deep learning CNN method is superior to the classical linear SVM on the EEG classification.