Ingrid Scholl

Challenges of medical image processing

In todays health care, imaging plays an important role throughout the entire clinical process from diagnostics and treatment planning to surgical procedures and follow up studies. Since most imaging modalities have gone directly digital, with continually increasing resolution, medical image processing has to face the challenges arising from large data volumes. In this paper, we discuss Kilo- to Terabyte challenges regarding (i) medical image management and image data mining, (ii) bioimaging, (iii) virtual reality in medical visualizations and (iv) neuroimaging. Due to the increasing amount of data, image processing and visualization algorithms have to be adjusted. Scalable algorithms and advanced parallelization techniques using graphical processing units have been developed. They are summarized in this paper. While such techniques are coping with the Kilo- to Terabyte challenge, the Petabyte level is already looming on the horizon. For this reason, medical image processing remains a vital field of research.

IR Stereo Kinect: Improving Depth Images by Combining Structured Light with IR Stereo

RGB-D sensors such as the Microsoft Kinect or the Asus Xtion are inexpensive 3D sensors. A depth image is computed by calculating the distortion of a known infrared light (IR) pattern which is projected into the scene. While these sensors are great devices they have some limitations. The distance they can measure is limited and they suffer from reflection problems on transparent, shiny, or very matte and absorbing objects. If more than one RGB-D camera is used the IR patterns interfere with each other. This results in a massive loss of depth information. In this paper, we present a simple and powerful method to overcome these problems. We propose a stereo RGB-D camera system which uses the pros of RGB-D cameras and combine them with the pros of stereo camera systems. The idea is to utilize the IR images of each two sensors as a stereo pair to generate a depth map. The IR patterns emitted by IR projectors are exploited here to enhance the dense stereo matching even if the observed objects or surfaces are texture-less or transparent. The resulting disparity map is then fused with the depth map offered by the RGB-D sensor to fill the regions and the holes that appear because of interference, or due to transparent or reflective objects. Our results show that the density of depth information is increased especially for transparent, shiny or matte objects.

Intuitive visual teleoperation for UGVs using free-look augmented reality displays

For numerous real-world applications teleoperated unmanned guided vehicles (UGVs) can quite successfully assist a human in fulfilling her mission objectives. It is important for the specialists to get an overview of the site quickly and with as intuitive means as possible. Our approach to an intuitive human-machine interface for visually teleoperating UGVs makes use of a spherical camera in combination with the Virtual Reality Head-Mounted Display (HMD) Oculus Rift. The Oculus Rift is equipped with an inertial measurement unit to track the teleoperators head orientation. With this orientation information, the current field of view is synthesized from the panoramic image coming from the spherical camera allowing a free look for the operator. In this paper, we present the hardware setup and the software system of our free-look HMD approach on our UGV. Our approach allows for multi-view, i.e, several operators can collaborate on a given task. What is more, we augment the spherical image with heading and orientation information of the robot as well as of other viewers. Our preliminary evaluation with a number of untrained user suggests that our free-look HMD offers an intuitive multi-view human-machine interface for teleoperating UGVs.

Comparing GPU-based multi-volume ray casting techniques

The most essential technique to visualize 3D scalar data is direct volume rendering. For many applications it is necessary that two or more 3D data are visualized simultaneously. We present an overview of data intermixing techniques for visualization with the direct volume rendering technique ray casting. The techniques are Classification Level Intermixing, Accumulation Level Intermixing and Image Level Intermixing. The algorithms are implemented on the Graphics Processing Unit (GPU) in order to be able to interact with the visualization in real-time. We use the new CUDA technology from Nvidia for that. We compare performance by measuring frames per seconds (FPS) and analyzing image quality with criteria contrast and depth effect. Depth effect is determined by a small user study. In most cases Accumulation Level Intermixing is the best choice.

Advances in medical image processing

For more than 20 years, the German Workshop on Medical Image Processing (Bildverarbeitung für die Medizin, BVM) is held annually and has been established recently as a European conference. In 2010, the workshop was held in Aachen, Germany. Based on a double-blind review process with at least three experts reviewing each manuscript, the best 20 papers have been invited contributing to this special issue. The thirteen submissions received have passed an international peer review process. Finally, eleven papers have been accepted for publication. This includes two invited papers by Scholl et al. and Wismueller.

Improving additive manufacturing by image processing and robotic milling

This paper describes an improvement of Additive Manufacturing (AM) to process products with complex inner and outer geometries such as turbine blades by using a combination of AM, optical measuring techniques and robotics. One problem of AM is the rough surface caused by the layer-by-layer 3D printing with granular material, so the geometries need to be post-processed with milling techniques. To overcome this problem, we implement an inline quality control management to post-process several manufactured layers by a integrated imaging and robotic techniques, especially while the geometries are still accessible. The produced surfaces are measured by an inline stereo vision system and are compared with the real 3D CAD model data. Detected differences are translated to movements of a robot arm which then follows the contours to smoothen the surfaces with a milling tool. We show our current state on the 3D imaging, on developing the driver for the deployed robot arm and on a 3D simulation using the Robot Operating System (ROS).

MQOne: Low-Cost Design for a Rugged-Terrain Robot Platform

Rugged terrain robot designs are important for field robotics missions. A number of commercial platforms are available, however, at an impressive price. In this paper, we describe the hardware and software component of a low-cost wheeled rugged-terrain robot. The robot is based on an electric children quad bike and is modified to be driven by wire. In terms of climbing properties, operation time and payload it can compete with some of the commercially available platforms, but at a far lower price.

Machine Vision System for Inline Inspection in Carbide Insert Production

In steps of the production chain of carbide inserts, such as unloading or packaging, the conformity test of the insert type is done manually, which causes a statistic increase of errors due to monotony and fatigue of the worker and the wide variety of the insert types. A machine vision system is introduced that captures digital frames of the inserts in the production line, analyses inspects automatically and measures four quality features: coating colour, edge radius, plate shape and chip-former geometry. This new method has been tested on several inserts of different types and has shown that the prevalent insert types can be inspected and robustly classified in real production environment and therefore improves the manufacturing automation.

Texture-based feature extraction using the wavelet transform on x rays

Focal bone lesions and bone tumors are of special interest in radiology because of their rare appearance (only one percent of all tumor diseases). This motivates a computer-assisted diagnosis recognizing bone tumors. Our image analysis extracts the radiomorphologic features in x rays using a texture-based approach. Diagnosing x rays, the radiologist examines regions of different size in x rays to gain both local and global impressions of the morphologic structure. In order to analyze the x ray in different resolutions, a multiresolution approach based on the wavelet transform is applied to the radiographs. To measure the informational content of the wavelet coefficients for the individual morphologic structures, we calculated a normalized summation of the absolute wavelet coefficients within a local N by N window and called this feature the local energy. We proved in different tests this feature and the parameter for calculating the wavelet transform for a correct classification of the medical structures, applying a topologic map from Kohonen. It is shown that the wavelet transform is well suited for the feature extraction of textures.

CRVM: Circular Random Variable-based Matcher - A Novel Hashing Method for Fast NN Search in High-dimensional Spaces.

Nearest Neighbour (NN) search is an essential and important problem in many areas, including multimedia databases, data mining and computer vision. For low-dimensional spaces a variety of tree-based NN search algorithms efficiently cope with finding the NN, for high-dimensional spaces, however, these methods are inefficient. Even for Locality Sensitive Hashing (LSH) methods which solve the task approximately by grouping sample points that are nearby in the search space into buckets, it is difficult to find the right parameters. In this paper, we propose a novel hashing method that ensures a high probability of NNs being located in the same hash buckets and a balanced distribution of data across all the buckets. The proposed method is based on computing a selected number of pairwise uncorrelated and uniformly-distributed Circular Random Variables (CRVs) from the sample points. The method has been tested on a large dataset of SIFT features and was compared to LSH and the Fast Library for Approximated NN search (FLANN) matcher with linear search as the base line. The experimental results show that our method significantly reduces the search query time while preserving the search quality, in particular for dynamic databases and small databases whose size does not exceed 200k points.