Search

Industrial X-ray analysis is common in aerospace, automotive or nuclear industries where structural integrity of some parts needs to be guaranteed. However, the interpretation of radiographic images is sometimes difficult and may lead to two experts disagree on defect classification. The automated defect recognition (ADR) system presented herein will reduce the analysis time and will also help reducing the subjective interpretation of the defects while increasing the reliability of the human inspector. Our convolutional neural network (CNN) model achieves 0.942 mAP@IoU = 0.50, which is considered as similar to expected human performance, when applied to an automotive aluminium castings dataset (GDXray). On an industrial environment, its inference time is less than 400 ms per 16 GB DICOM image (16 bits), so it can be installed on production facilities with no impact on delivery time. In addition, an ablation study of the main hyper-parameters to optimise model accuracy from the initial baseline result of 0.75 mAP up to 0.942 mAP, was also conducted. [ABSTRACT FROM AUTHOR]

Castings produced for the automotive industry are considered important components for overall roadworthiness. To ensure the safety of construction, it is necessary to check every part thoroughly using non-destructive testing. X-ray testing rapidly became the accepted way of controlling the quality of die-cast pieces. In this paper, we present a Convolutional Neural Network (CNN) for defect detection in castings. In order to train the CNN model, a large dataset is necessary. We build the dataset by using synthetic defects. They are simulated using 3D ellipsoidal models and Generative Adversarial Networks (GAN). We compare different portions of ellipsoidal/GAN defects in the training subset. In our experiments, the use of GAN defects does not play a relevant role in this solution. However, ellipsoidal defects helped to achieve better performance. Ellipsoidal defects from any size and orientation could be superimposed onto real X-ray images in any location. In addition, we tested several CNN configurations, the best one, that we call Xnet-II, has 30 layers and more than 1,350,000 parameters. It has been trained using a dataset with around 640,000 patches containing 50% of ellipsoidal defects and 50% of real background captured from different casting types. The model was tested using sliding-windows methodology on whole X-ray images achieving promising results (mPA = 0.7102): the model was able to detect real defects from different casting types. We believe that the methodology presented could be used in similar projects that have to deal with automated detection of defects. [ABSTRACT FROM AUTHOR]

In this paper, we describe an ANN with a modified performance function which is used in an automatic inspection system of welding defects in radiographic images. In a first stage, image processing techniques, including noise reduction, contrast enhancement, thresholding and labelling, were implemented to help in the recognition of weld regions and the detection of weld defects. In a second stage, a set of geometrical features which characterise the defect shape and orientation was proposed and extracted between defect candidates. In a third stage, an artificial neural network for weld defect classification was used under a regularisation process with different architectures for the input layer and the hidden layer. Our aim is to analyse this ANN modifying the performance function using a γ parameter in its function, for different neurons in the input and hidden layer in order to obtain a better performance on the classification stage. The automatic system of recognition and classification proposed consists in detecting the four main types of weld defects met in practice plus the non-defect type. The results was compared with the aim to know the parameters that allow the best classification. The correlation coefficients, confusion matrix and the accuracy or the proportion of the total number of predictions that were correct was determined obtaining a value of 80% for the ANN using a modified performance function with a parameter γ=0.6. [ABSTRACT FROM AUTHOR]

The creation of a non-destructive technique that enables the automatic detection of defects is desirable, and TOFD (Time-Of-Flight Diffraction) technique is gaining rapid prominence due to its high accuracy in detecting, positioning and sizing flaws in steel structures. In this type of imaging, cracks are characterized using sets of hyperbolas, where summit positions correspond to crack tip positions. However, ultrasonic diffracted signals are often too low and difficult to distinguish from noise, and when large structures are inspected, the quantity of data can be extremely large, with the area of interest being very small in comparison to the image size. This paper describes a method that avoids the image formation, replacing it with a sparse matrix (as there is no reason to store and operate on an excessive number of zeros), and automates crack detection by analyzing the curve formed by the sparse matrix elements. The sparse matrix is formed using Split-Spectrum Processing, which enhances the signal-to-noise ratio. The Randomized Hough transform is then applied on the sparse matrix elements to detect the hyperbolas that characterize the crack defects. [ABSTRACT FROM AUTHOR]