Search

Background: Hyperspectral imaging techniques have emerged as powerful tools for non-invasive investigation of artworks. This paper employs either reflectance imaging spectroscopy (RIS) or macroscopic X-ray fluorescence (MA-XRF) imaging in combination with macroscopic X-ray powder diffraction (MA-XRPD) for state-of-the-art chemical imaging of painted cultural heritage artefacts. While RIS can provide molecular information and MA-XRF can offer elemental distribution maps of paintings of high lateral resolution, the unique advantage of MA-XRPD lies in its ability to visualize the distributions of specific pigments and estimate in a quantitative manner the relative concentrations of the crystalline phases at the surface of artworks. However, MA-XRPD is more time-consuming and offers a lower lateral resolution than RIS and MA-XRF., Results: This study introduces a machine learning (ML) approach to obtain the distribution of specific compounds on the surface of artworks with a resolution that is comparable to that of RIS and MA-XRF data but with the compound specificity of MA-XRPD. The general aim is to expedite non-destructive artwork imaging analysis by fusing data from different imaging modalities via machine learning models. The effect of preprocessing techniques to enhance the predictive accuracy of the models is explored. The paper demonstrates the method's efficacy on a 16th-century illuminated manuscript, showcasing the feasibility of predicting compound-specific distribution maps. Three evaluation methods-visual examination of the predicted distribution, root mean square errors (RMSE), and feature permutation importance (FPI)-are employed to assess model performance. Fusing MA-XRF with MA-XRPD led to the best RMSE scores overall. However, fusing the RIS and MA-XRPD data blocks also yield very satisfactory and easily interpretable high-resolution compound maps., Significance: While MA-XRPD allows for highly specific imaging of artworks, its time-consuming nature and limited resolution presents a bottleneck during non-invasive imaging of painted works of art. By integrating data from more time-efficient hyperspectral techniques such as MA-XRF and RIS, and employing machine learning, we expedite the process without compromising accuracy. The fusion process can also denoise the distribution maps, improving their readability for heritage professionals and art historical scholars., Competing Interests: Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:Arthur Gestels reports equipment, drugs, or supplies and writing assistance were provided by Netherlands National Museum. Arthur Gestels reports writing assistance was provided by Scientific Research Department at the National Gallery of Art, Washington, DC. Arthur Gestels reports financial support was provided by Research Foundation Flanders. Arthur Gestels reports financial support was provided by Interreg Flanders-Netherlands. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

The high incidence of oil spills in port areas poses a serious threat to the environment, prompting the need for efficient detection mechanisms. Utilizing automated drones for this purpose can significantly improve the speed and accuracy of oil spill detection. Such advancements not only expedite cleanup operations, reducing environmental harm but also enhance polluter accountability, potentially deterring future incidents. Currently, there's a scarcity of datasets employing RGB images for oil spill detection in maritime settings. This paper presents a unique, annotated dataset aimed at addressing this gap, leveraging a neural network for analysis on both desktop and edge computing platforms. The dataset, captured via drone, comprises 1268 images categorized into oil, water, and other, with a convolutional neural network trained using an Unet model architecture achieving an F1 score of 0.71 for oil detection. This underscores the dataset's practicality for real-world applications, offering crucial resources for environmental conservation in port environments., (© 2024. The Author(s).)

This study presents a novel method for the detection and quantification of atmospheric corrosion products on carbon steel. Using hyperspectral imaging (HSI) in the short-wave infrared range (SWIR) (900-1700 nm), we are able to identify the most common corrosion minerals such as: α-FeO(OH) (goethite), γ-FeO(OH) (lepidocrocite), and γ-Fe 2 O 3 (maghemite). Six carbon steel samples were artificially corroded in a salt spray chamber, each sample with a different duration (between 1 h and 120 hours). These samples were analysed by scanning X-ray diffraction (XRD) and also using a SWIR HSI system. The XRD data is used as baseline data. A random forest regression algorithm is used for training on the combined XRD and HSI data set. Using the trained model, we can predict the abundance map based on the HSI images alone. Several image correlation metrics are used to assess the similarity between the original XRD images and the HSI images. The overall abundance is also calculated and compared for XRD and HSI images. The analysis results show that we are able to obtain visually similar images, with error rates ranging from 3.27 to 13.37%. This suggests that hyperspectral imaging could be a viable tool for the study of corrosion minerals., Competing Interests: There are no conflicts to declare., (This journal is © The Royal Society of Chemistry.)

In this study, we propose a new method to identify corrosion minerals in carbon steel using hyperspectral imaging (HSI) in the shortwave infrared range (900-1700 nm). Seven samples were artificially corroded using a neutral salt spray test and examined using a hyperspectral camera. A normalized cross-correlation algorithm is used to identify four different corrosion minerals (goethite, magnetite, lepidocrocite and hematite), using reference spectra. A Fourier Transform Infrared spectrometer (FTIR) analysis of the scraped corrosion powders was used as a ground truth to validate the results obtained by the hyperspectral camera. This comparison shows that the HSI technique effectively detects the dominant mineral present in the samples. In addition, HSI can also accurately predict the changes in mineral composition that occur over time.

In this article, we report the use of a Confocal Laser Scanning Microscope (CLSM) to apply a qualitative assessment of atmospheric corrosion on steel samples. From the CLSM, we obtain high-resolution images, together with a 3D heightmap. The performance of four different segmentation algorithms that use the high-resolution images as input is qualitatively assessed and discussed. A novel 3D segmentation algorithm based on the shape index is presented and compared to the 2D segmentation algorithms. From this analysis, we conclude that there is a significant difference in performance between the 2D segmentation algorithms and that the 3D method can be an added value to the detection of corrosion.