Techn. Fakultät Willkommen am Institut für Informatik FAU-Logo

AmirAbbas Davari M. Sc.

Researcher in the Computer Vision (CV) group at the Pattern Recognition Lab of the Friedrich-Alexander-Universität Erlangen-Nürnberg

Old Master Drawings' Sketch layer decomposition using hyperspectral image analysis

Recently, the progress in computer science made links between this field and art history possible. The examination of cultural artifacts particularly gained new insights in materials and object structure from new possibilities in high-resolution image acquisition and data processing. However, although there is a large number of specialized examination methods (for example IR/UV reflectography, fluorescence, micro-XRF, Raman spectroscopy), the research on hand drawings still suffers from major diagnostic gaps: due to the particular properties of red chalk, it is currently not possible to visualize preparatory drawings from red chalk if they have been overlaid by inks.

Up until the end of the 19th century, hand drawings were typically created with different materials in several working steps. In the final drawing, multiple layers of different materials overlap, such that the lower, older layers can oftentimes not be identified visually. However, these layers represent the various steps of the genesis of the respective work. Thus, separating these layers by means of imaging and image processing promises a direct look into the artistic creation process of the work. Therefore, such a technical approach may supply answers to key questions of art history on the object and the artist, and may help for objectivation of attribution and authenticity.

In this project, we investigate approaches to close this diagnostic gap. In contrast to image acquisitions with a limited spectral window (like infrared or ultraviolet light), we propose to image an object using a multispectral camera and to process the acquired data with methods from the field of pattern recognition. A multispectral camera operates primarily in the range of visible light but subdivides the light into much more channels than red, green, and blue. This allows differentiating materials of different physical compositions from their reflection patterns after processing the data with a computational algorithm. We evaluate our proposed methods on drawings that were created to exactly mimic the original work process, using the same materials and papers. The controlled creation process provides knowledge about the drawing layers, which allows assessing the accuracy of the method. This approach has the potential to avoid radiation damage to the work of art under examination, while still being able to provide meaningful information. Thus, it appears as a method that can have broad applications in the domain of art history and conservation of art that extend beyond the range of currently used approaches.



Ink/Chalk layers of a Phantom Drawing Image
Hyperspectral Remote Sensing Image Analysis

Remote sensing is nowadays of paramount importance for several application fields, including environmental monitoring, urban planning, ecosystem-oriented natural resources management, urban change detection and agricultural region monitoring. Majority of the aforementioned monitoring and detection applications requires at some stage a label map of the remotely sensed images, where individual pixels are marked as members of specific classes, e.g. water, asphalt, grass, etc. In other words, classification is a crucial step for several remote sensing applications. It is widely acknowledged that exploiting both the spectral as well as spatial properties of pixels, improves classification performance with respect to using only spectral based features.

In this regard, morphological profiles (MP) are one of the popular and powerful image analysis techniques that enable us to compute such spectral-spatial pixel descriptions. They have been studied extensively in the last decade and their effectiveness has been validated repeatedly.

The characterization of spatial information obtained by the application of a MP is particularly suitable for representing the multi-scale variations of image structures, but they are limited by the shape of the structuring elements. To avoid this limitation, morphological attribute profiles (AP) have been developed. By operating directly on connected components instead of pixels, not only we are able to employ arbitrary region descriptors (e.g. shape, color, texture, etc.) but it paves the way for object based image analysis as well. In addition, APs can be implemented efficiently by means of hierarchical image representations, e.g. Max-/Min-tree and alpha-tree.

Although the aforementioned descriptors served well for spectral-spatial image description, they have their own bottlenecks. In this project, we investigate, study and  propose new descriptors and methods for both spectrally and spatially describing the hyperspectral remote sensing images.



Sample hyperspectral remote sensing image classification framework. Image is from our SIU 2015 Paper.