Ronny Hänsch

Automated drop detection using image analysis for online particle size monitoring in multiphase systems

Abstract Image analysis has become a powerful tool for the work with particulate systems, occurring in chemical engineering. A major challenge is still the excessive manual work load which comes with such applications. Additionally manual quantification also generates bias by different observers, as shown in this study. Therefore a full automation of those systems is desirable. A MATLAB ® based image recognition algorithm has been implemented to automatically count and measure particles in multiphase systems. A given image series is pre-filtered to minimize misleading information. The subsequent particle recognition consists of three steps: pattern recognition by correlating the pre-filtered images with search patterns, pre-selection of plausible drops and the classification of these plausible drops by examining corresponding edges individually. The software employs a normalized cross correlation procedure algorithm. The program has reached hit rates of 95% with an error quotient under 1% and a detection rate of 250 particles per minute depending on the system.

Complex-Valued Multi-Layer Perceptrons – An Application to Polarimetric SAR Data

Multi-Layer Perceptrons (MLPS) are powerful function approximators. In the last decades they were successfully applied to many different regression and classification problems. Their characteristics and convergence properties are well studied and relatively well understood, but they were originally designed to work with real-valued data. The main focus of this paper is the classification of polarimetric synthetic aperture radar (POLSAR) data which are a complex-valued signal. Instead of using an arbitrarily projection of this complex-valued data to the real domain, the paper proposes the usage of complex-valued MLPS (CV-MLPS), which are an extension of MLPS to the complex domain. The paper provides a detailed yet general derivation of the complex backpropagation algorithm and mentions related problems as well as possible solutions. Furthermore, it evaluates the performance of CV-MLPS in a land-cover classification task in POLSAR images under several learning conditions, and compares the proposed classifier with standard methods. The experimental results show that CV-MLPS are successfully applicable to classification tasks in POLSAR data. They show good convergence properties and a better performance if compared to real-valued MLPS.

Integrated Matching and Geocoding of SAR and Optical Satellite Images

In this paper matching of SAR and optical remote sensing images in object space is investigated. First, the matching performance of descriptors and localizers for point features extracted from preprocessed images is analyzed. In particular, the increase in matching performance when making use of approximate localization information is documented. Secondly, the combined geocoding of SAR and optical image data using both orbit information and digital surface models as well as matched points is introduced and investigated using a radar-photogrammetric least-squares adjustment approach.

Random Forests for building detection in polarimetric SAR data

Building detection from Synthetic Aperture Radar (SAR) images states a particular important as well as difficult problem. The high-resolution which is necessary to distinguish single buildings as well as the geometric and di-electric properties of dense urban areas cause most assumptions to fail, that are commonly made in SAR data analysis. This paper proposes the usage of Random Forests for building detection from high-resolution Polarimetric Synthetic Aperture Radar (PolSAR) imagery. Random Forests can handle high-dimensional input and therefore a large set of different features, they are known to lead to good classification performance in terms of robustness and accuracy, and are nevertheless seldomly applied to analysis of PolSAR images in general and building detection in particular. This paper presents first results of Random Forests when applied to a building detection task and shows their successful applicability.

Object Recognition from Polarimetric SAR Images

In general, object recognition from images is concerned with separating a connected group of object pixels from background pixels and identifying or classifying the object. The indication of the image area covered by the object makes information which is implicitly given by the group of pixels, explicit by naming the object. The implicit information can be contained in the measurement values of the pixels or in the locations of the pixels relative to each other. While the former represent radiometric properties, the latter is of geometric nature describing the shape or topology of the object.

Semi-supervised learning for classification of polarimetric SAR-data

Supervised learning algorithms are important methods to automatically interpret image data in general as well as PolSAR data in particular. However, they suffer from the need of a training set, which has to contain manually labelled data. Un-supervised methods do not demand this kind of data, but cannot be directly used to assign user-defined class labels to image regions. This paper proposes a semi-supervised method to overcome both shortcomings. The data is analysed by an un-supervised clustering algorithm under the usage of all available information. Simultaneously each pixel is classified by a supervised method using the information available at the current phase of clustering.

Automated image analysis for trajectory determination of single drop collisions

Abstract The fundamental analysis of drop coalescence probability in liquid/liquid systems is necessary to reliably predict drop size distributions in technical applications. For this crucial investigation two colliding oil drops in continuous water phase were recorded with different high speed camera set-ups under varying conditions. In order to analyze the huge amount of recorded image sequences with varying resolutions and qualities, a robust automated image analysis was developed. This analysis is able to determine the trajectories of two colliding drops as well as the important events of drop detachment from cannulas and their collision. With this information the drop velocity in each sequence is calculated and mean values of multiple drop collisions are determined for serial examinations of single drop collisions. Using the developed automated image analysis for drop trajectory and velocity calculation, approximately 1–2 recorded high speed image sequences can be evaluated per minute.

GRAPH-CUT SEGMENTATION OF POLARIMETRIC SAR IMAGES

Segmentation of Synthetic Aperture Radar (SAR) images is often only understood as the partitioning of the image into rather small regions which are homogeneous with respect to scattering processes. This paper proposes an adaption of the graph-cut image segmentation framework to the unique characteristics of polarimetric SAR images by using a Wishart-distribution based distance measure for local segmentation cues and simple, real-valued features derived from the complex-valued coherency matrix. The proposed method is evaluated on different polarimetric SAR images, for different objects of interest, and with a wide range of parameters. The results show that the proposed framework is able to derive accurate object/non-object segmentations. Best results are obtained for forest areas by usage of a log-transform of the polarimetric intensities.

Classification of PolSAR Images by Stacked Random Forests

This paper proposes the use of Stacked Random Forests (SRF) for the classification of Polarimetric Synthetic Aperture Radar images. SRF apply several Random Forest instances in a sequence where each individual uses the class estimate of its predecessor as an additional feature. To this aim, the internal node tests are designed to work not only directly on the complex-valued image data, but also on spatially varying probability distributions and thus allow a seamless integration of RFs within the stacking framework. Experimental results show that the classification performance is consistently improved by the proposed approach, i.e., the achieved accuracy is increased by 4 % and 7 % for one fully- and one dual-polarimetric dataset. This increase only comes at the cost of a linear increased training and prediction time, which is rather limited as the method converges quickly.

Machine-learning based detection of corresponding interest points in optical and SAR images

One of the major problems of keypoint-based alignment of SAR and optical images is that keypoint operators react to very different object structures in both image types. This leads to a small mutual overlap in the corresponding sets of keypoints. This paper proposes to cast the task of keypoint detection as a classification problem. A machine-learning based classifier is trained to predict whether a SAR image pixel corresponds to a keypoint in the optical image or not. Experimental results indicate, that the mutual overlap of keypoints can be doubled by the proposed approach.