No abstract

This paper attempts to describe a pipeline architecture synthesis tool dedicated to signal processing applications. This approach relies on the use of a design strategy and of a generic architectural model, using optimized control of resources. GAUTi takes a VHDL description of an application as input, and generates the optimal structural and functional VHDL description of a dedicated architecture. The results obtained by GAUT are intended for an application in acoustic echo cancellation

The accurate detection of cracks in paintings, which generally portray rich and varying content, is a challenging task. Traditional crack detection methods are often lacking on recent acquisitions of paintings as they are poorly adapted to high-resolutions and do not make use of the other imaging modalities often at hand. Furthermore, many paintings portray a complex or cluttered composition, significantly complicating a precise detection of cracks when using only photographic material. In this paper, we propose a fast crack detection algorithm based on deep convolutional neural networks (CNN) that is capable of combining several imaging modalities, such as regular photographs, infrared photography and X-Ray images. Moreover, we propose an efficient solution to improve the CNN-based localization of the actual crack boundaries and extend the CNN architecture such that areas where it makes little sense to run expensive learning models are ignored. This allows us to process large resolution scans of paintings more efficiently. The proposed on-line method is capable of continuously learning from newly acquired visual data, thus further improving classification results as more data becomes available. A case study on multimodal acquisitions of the Ghent Altarpiece, taken during the currently ongoing conservation-restoration treatment, shows improvements over the state-of-the-art in crack detection methods and demonstrates the potential of our proposed method in assisting art conservators.INDEX TERMS Digital painting analysis, crack detection, virtual restoration, machine learning, morphological filtering, convolutional neural networks, transfer learning, multimodal data, Ghent Altarpiece.

A combination of large-scale and micro-scale elemental imaging, yielding elemental distribution maps obtained by, respectively non-invasive macroscopic X-ray fluorescence (MA-XRF) and by secondary electron microscopy/energy dispersive X-ray analysis (SEM-EDX) and synchrotron radiation-based micro-XRF (SR μ-XRF) imaging was employed to reorient and optimize the conservation strategy of van Eyck's renowned Ghent Altarpiece. By exploiting the penetrative properties of X-rays together with the elemental specificity offered by XRF, it was possible to visualize the original paint layers by van Eyck hidden below the overpainted surface and to simultaneously assess their condition. The distribution of the high-energy Pb-L and Hg-L emission lines revealed the exact location of hidden paint losses, while Fe-K maps demonstrated how and where these lacunae were filled-up using an iron-containing material. The chemical maps nourished the scholarly debate on the overpaint removal with objective, chemical arguments, leading to the decision to remove all skillfully applied overpaints, hitherto interpreted as work by van Eyck. MA-XRF was also employed for monitoring the removal of the overpaint during the treatment phase. To gather complementary information on the in-depth layer build-up, SEM-EDX and SR μ-XRF imaging was used on paint cross sections to record micro-scale elemental maps.