Aleksandra Sima

Optimizing SIFT for Matching of Short Wave Infrared and Visible Wavelength Images

The scale invariant feature transform (SIFT) is a widely used interest operator for supporting tasks such as 3D matching, 3D scene reconstruction, panorama stitching, image registration and motion tracking. Although SIFT is reported to be robust to disparate radiometric and geometric conditions in visible light imagery, using the default input parameters does not yield satisfactory results when matching imagery acquired at non-overlapping wavelengths. In this paper, optimization of the SIFT parameters for matching multi-wavelength image sets is documented. In order to integrate hyperspectral panoramic images with reference imagery and 3D data, corresponding points were required between visible light and short wave infrared images, each acquired from a slightly different position and with different resolutions and geometric projections. The default SIFT parameters resulted in too few points being found, requiring the influence of five key parameters on the number of matched points to be explored using statistical techniques. Results are discussed for two geological datasets. Using the SIFT operator with optimized parameters and an additional outlier elimination method, allowed between four and 22 times more homologous points to be found with improved image point distributions, than using the default parameter values recommended in the literature.

Computer-aided image geometry analysis and subset selection for optimizing texture quality in photorealistic models

Photorealistic 3D models are used for visualization, interpretation and spatial measurement in many disciplines, such as cultural heritage, archaeology and geoscience. Using modern image- and laser-based 3D modelling techniques, it is normal to acquire more data than is finally used for 3D model texturing, as images may be acquired from multiple positions, with large overlap, or with different cameras and lenses. Such redundant image sets require sorting to restrict the number of images, increasing the processing efficiency and realism of models. However, selection of image subsets optimized for texturing purposes is an example of complex spatial analysis. Manual selection may be challenging and time-consuming, especially for models of rugose topography, where the user must account for occlusions and ensure coverage of all relevant model triangles. To address this, this paper presents a framework for computer-aided image geometry analysis and subset selection for optimizing texture quality in photorealistic models. The framework was created to offer algorithms for candidate image subset selection, whilst supporting refinement of subsets in an intuitive and visual manner. Automatic image sorting was implemented using algorithms originating in computer science and information theory, and variants of these were compared using multiple 3D models and covering image sets, collected for geological applications. The image subsets provided by the automatic procedures were compared to manually selected sets and their suitability for 3D model texturing was assessed. Results indicate that the automatic sorting algorithms are a promising alternative to manual methods. An algorithm based on a greedy solution to the weighted set-cover problem provided image sets closest to the quality and size of the manually selected sets. The improved automation and more reliable quality indicators make the photorealistic model creation workflow more accessible for application experts, increasing the users confidence in the final textured model completeness.

Spatially variable filters — Expanding the spectral dimension of compact cameras for remotely piloted aircraft systems

Recent advances in hyperspectral imaging techniques using spectral filters deposited directly onto an image sensor chip, and suitable for RPAS platforms, are reported in this paper. New filter configurations for a compact spectral camera are described and a prototype of the compact hyperspectral payload developed for small RPAS systems, as well as the first data acquired with the new camera are presented. The spectral range of the payload was optimized for earth observations such as vegetation monitoring or water quality studies. Although the spatially variable filters have to date only been used in small satellite sensors, this technology has a clear potential for RPAS platforms.

Information measures for terrain visualization

Many quantitative and qualitative studies in geoscience research are based on digital elevation models (DEMs) and 3D surfaces to aid understanding of natural and anthropogenically-influenced topography. As well as their quantitative uses, the visual representation of DEMs can add valuable information for identifying and interpreting topographic features. However, choice of viewpoints and rendering styles may not always be intuitive, especially when terrain data are augmented with digital image texture. In this paper, an information-theoretic framework for object understanding is applied to terrain visualization and terrain view selection. From a visibility channel between a set of viewpoints and the component polygons of a 3D terrain model, we obtain three polygonal information measures. These measures are used to visualize the information associated with each polygon of the terrain model. In order to enhance the perception of the terrains shape, we explore the effect of combining the calculated information measures with the supplementary digital image texture. From polygonal information, we also introduce a method to select a set of representative views of the terrain model. Finally, we evaluate the behaviour of the proposed techniques using example datasets. A publicly available framework for both the visualization and the view selection of a terrain has been created in order to provide the possibility to analyse any terrain model. HighlightsInformation theory framework for terrain visualization.Polygonal information is used to enhance the terrain visualization.N-best view algorithm based on viewpoint quality measures.Publicly available framework.

Radiometric calibration of the cosi hyperspectral RPAS camera

The COSI hyperspectral imaging system, suitable for small RPAS, is able to produce high resolution hyperspectral data products. By extensive inflight testing, we have identified the main challenges for achieving reliable high quality results. Based on these insights, we propose a refined radiometric calibration strategy. It uses a set of three reference targets, two grey and one colored target, which are to be measured inflight. We present on-ground measurements of the targets with COSI, as in flight measurements, demonstrating the merits of the approach are still ongoing.

Linear variable filters — A camera system requirement analysis for hyperspectral imaging sensors onboard small Remotely Piloted Aircraft Systems

Improving the spectral detail of earth observation imaging from Remotely Piloted Aircraft Systems (RPAS) can greatly expand its potential for use in vegetation monitoring and specifically in precision agriculture. Spatially variable interference filters which can be placed very close to the image sensor offer an excellent opportunity for reducing the size, mass and complexity of hyperspectral imagers, allowing them to be mounted onboard small RPAS. Recent advances in filter deposition techniques allow to directly deposit interference filters on an image sensor. The monolithic integration of optical hyperspectral filters on top of a standard CMOS image sensor has been demonstrated by IMEC. Compared to the more conventional deposition of filters onto an external glass substrate, this new approach offers advantages in terms of cost, alignment accuracy, straylight, etc. A hyperspectral camera prototype compatible with small RPAS has been developed by VITO to demonstrate the potential of LVF-based compact spectral cameras. Whereas application of the filter technology offers major advantages for RPAS systems, it still faces some important challenges. The prototype system specifications need to fit a fixed wing RPAS platform that is able to cover several km2 in a single flight with hyperspectral geo-information. It remains challenging to make a sufficiently compact camera system, achieve precise spectral band registration, handle the amount of data to be processed and cope with limited integration times possible during acquisition.