Harald Grossauer

A Combined PDE and Texture Synthesis Approach to Inpainting

While there is a vast amount of literature considering PDE based inpainting and inpainting by texture synthesis, only a few publications are concerned with combination of both approaches. We present a novel algorithm which combines both approaches and treats each distinct region of the image separately. Thus we are naturally lead to include a segmentation pass as a new feature. This way the correct choice of texture samples for the texture synthesis is ensured. We propose a novel concept of “local texture synthesis” which gives satisfactory results even for large domains in a complex environment.

Using the complex Ginzburg-Landau equation for digital inpainting in 2D and 3D

Recently, several different approaches for digital inpainting have been proposed in the literature. We give a review and introduce a novel approach based on the complex Ginzburg-Landau equation. The use of this equation is motivated by some of its remarkable analytical properties. While common inpainting technology is especially designed for restorations of two dimensional image data, the Ginzburg-Landau equation can straight forwardly be applied to restore higher dimensional data, which has applications in frame interpolation, improving sparsely sampled volumetric data and to fill in fragmentary surfaces. The latter application is of importance in architectural heritage preservation. We discuss a stable and efficient scheme for the numerical solution of the Ginzburg-Landau equation and present some numerical experiments. We compare the performance of our algorithm with other well established methods for inpainting.

Photoacoustic microtomography using optical interferometric detection

A device for three-dimensional (3-D) photoacoustic tomography with resolution in the range of tens of micrometers is presented that uses a light beam for interferometric detection of acoustic waves. Reconstruction of the 3-D initial pressure distribution from the signals representing line integrals of the acoustic field is a two-step process. It uses an inversion of 2-D wave propagation to obtain line projections of the initial pressure distribution and the inverse Radon transform. The light beam, propagating freely in a water bath, is scanned either in an arc- or box-shaped curve around the object. Simulations are performed to compare the two scanning procedures. The projection images are obtained either using the filtered back projection algorithm for the pi-arc scanning mode or the frequency domain algorithm for the box scanning mode. While the former algorithm provides slightly better image quality, the latter is about 20 times faster. The ability of the photoacoustic tomography device to create 3-D images with constant resolution throughout the reconstruction volume is demonstrated experimentally using a human hair phantom. These measurements revealed a 3-D resolution below 100 mum. In a second experiment, 3-D imaging of an isolated mouse heart is demonstrated to show the applicability for preclinical and biological research.

Analysis of Iterative Methods for Solving a Ginzburg-Landau Equation

Very recently we have proposed to use a complex Ginzburg-Landau equation for high contrast inpainting, to restore higher dimensional (volumetric) data (which has applications in frame interpolation), improving sparsely sampled data and to fill in fragmentary surfaces. In this paper we review digital inpainting algorithms and compare their performance with a Ginzburg-Landau inpainting model. For the solution of the Ginzburg-Landau equation we compare the performance of several numerical algorithms. A stability and convergence analysis is given and the consequences for applications to digital inpainting are discussed.

Photoacoustic tomography of ex vivo mouse hearts with myocardial infarction

In the present study, we evaluated the applicability of ex vivo photoacoustic imaging (PAI) on small animal organs. We used photoacoustic tomography (PAT) to visualize infarcted areas within murine hearts and compared these data to other imaging techniques [magnetic resonance imaging (MRI), micro-computed tomography] and histological slices. In order to induce ischemia, an in vivo ligation of the left anterior descending artery was performed on nine wild-type mice. After varying survival periods, the hearts were excised and fixed in formaldehyde. Samples were illuminated with nanosecond laser pulses delivered by a Nd:YAG pumped optical parametric oscillator. Ultrasound detection was achieved using a Mach-Zehnder interferometer (MZI) working as an integrating line detector. The voxel data were computed using a Fourier-domain based reconstruction algorithm, followed by inverse Radon transforms. The results clearly showed the capability of PAI to visualize myocardial infarction and to produce three-dimensional images with a spatial resolution of approximately 120 μm. Regions of affected muscle tissue in PAI corresponded well with the results of MRI and histology. Photoacoustic tomography utilizing a MZI for ultrasound detection allows for imaging of small tissue samples. Due to its high spatial resolution, good soft tissue contrast and comparatively low cost, PAT offers great potentials for imaging.

Photoacoustic tomography of pathological tissue in ex-vivo mouse hearts

In the present study, we evaluate the applicability of ex-vivo photoacoustic imaging (PAI) in organs of small animals. We used photoacoustic tomography (PAT) to visualize infarcted areas within mouse hearts and compared it to other imaging techniques (MRI and μCT). In order to induce ischemia an in-vivo ligation of the Ramus interventricularis anterior (RIVA, left anterior descending, LAD) was performed on nine wild type C41 mice. After varying survival periods the mice were sacrificed. The hearts were excised and immediately transferred into a formaldehyde solution for conservation. Various wavelengths in the visible and near infrared region (500 nm - 1000 nm) had been tested to find the best representation of the ischemic regions. Samples were illuminated with nanosecond laser pulses delivered by an Nd:YAG pumped optical parametric oscillator. Ultrasound detection was achieved by an optical Mach-Zehnder interferometer working as an integrating line detector. For acoustic coupling the samples were located inside a water tank. The voxel data are computed from the measurement data by a Fourier-domain based reconstruction algorithm, followed by a sequence of inverse Radon transforms. Results clearly show the capability of PAI to detect pathological tissue and the possibility to produce three-dimensional images with resolutions well below 100 μm. Different wavelengths allow the representation of structure inside an organ or on the surface even without contrast enhancing tracers.

Photoacoustic microtomography: system characterization and first results on biological specimens

A device for three-dimensional (3D) photoacoustic tomography with resolution in the range of tens of micrometers is presented. It is based on a focused laser beam as detector, which propagates near the imaging object in a water bath and is part of a Mach-Zehnder interferometer. By scanning the beam relative to the object, data for 3D image reconstruction are acquired. Using a phantom consisting of human hairs, the resolution of the setup is demonstrated. Using a flea and an excised mouse heart as samples, the imaging capabilities of the device for complex biological objects are shown.