Showing papers by "Sidharta Gautama published in 2010"

Geodatabase information processing

Abstract: A system is described for processing data of at least one location aware devices for obtaining geo-information. The system comprises an input means for obtaining vector-based data from at least one location aware devices, a data processor for inserting information from the vector-based data into a raster-based data structure so as to derive geo-information based on the raster-based data structure. A corresponding method as well as corresponding computer program products also are described.

On the accuracy of 3D landscapes from UAV image data

Abstract: Our surroundings change all the time. Applications that require 3D models of a changing terrain, such as urban planning, are becoming ever more demanding with respect to the cost to create them and the accuracy of the result. A novel, cheap and fast solution for this problem is given by a UAV to take aerial images of the terrain in question, in combination with structure from motion algorithms to create a 3D model from those aerial images. However the question remains whether these on-the-fly 3D maps can match the accuracy of classical surveyor based models, which require more time to create. In this paper we investigate this question, and find that under certain conditions the accuracy of the UAV based model matches the accuracy of surveyor generated measurements.

Segmentation quality evaluation for large scale mapping purposes in Flanders, Belgium

Abstract: In Flanders the large scale reference database called GRB, takes care of the layout, exchange and management of large scale geographic information with respect to, amongst others, roads, buildings and parcels. As Flanders is extremely urbanized (average population density of about 450 inhabitants per square kilometer), the large scale maps need to be highly accurate. Currently, accuracies at the centimeter level are guaranteed due to topographic field measurements aided by standard photogrammetry based on analogue aerial photographs. In order to speed up the GRB production and to ensure large scale map products at the long term, it is essential to automate this labour-intensive, but highly accurate production process. Segmentation of very high resolution digital images could be an alternative approach for maintaining and updating the Flemish GRB as long as high accuracy segmentation results are obtained. Based on DMC images (8 cm ground resolution) and several reference buildings, a comprehensive sensitivity analysis is performed testing different segmentation parameter settings in order to gain insight into their impact on segmentation accuracy. The segmentation quality is evaluated using similarity measures focusing on aspects of presence, shape and positional accuracy where emphasis is placed on interpretability of the measures with respect to operational conditions put on the reference data. The end user should be able to read the measures and link this to the return-on-investment he will gain by using a given segmentation process on his data.

Automatic identification of Caenorhabditis elegans in population images by shape energy features

Abstract: Experiments on model organisms are used to extend the understanding of complex biological processes. In Caenorhabditis elegans studies, populations of specimens are sampled to measure certain morphological properties and a population is characterized based on statistics extracted from such samples. Automatic detection of C. elegans in such culture images is a difficult problem. The images are affected by clutter, overlap and image degradations. In this paper, we exploit shape and appearance differences between C. elegans and non-C. elegans segmentations. Shape information is captured by optimizing a parametric open contour model on training data. Features derived from the contour energies are proposed as shape descriptors and integrated in a probabilistic framework. These descriptors are evaluated for C. elegans detection in culture images. Our experiments show that measurements extracted from these samples correlate well with ground truth data. These positive results indicate that the proposed approach can be used for quantitative analysis of complex nematode images.

Speeding Up Structure from Motion on Large Scenes Using Parallelizable Partitions

Abstract: Structure from motion based 3D reconstruction takes a lot of time for large scenes which consist of thousands of input images. We propose a method that speeds up the reconstruction of large scenes by partitioning it into smaller scenes, and then recombining those. The main benefit here is that each subscene can be optimized in parallel. We present a widely usable subdivision method, and show that the difference between the result after partitioning and recombination, and the state of the art structure from motion reconstruction on the entire scene, is negligible.

Automating actualisation of geographic information using the twofold fuzzy region model

Abstract: In this paper a new method is proposed for automating quality control and actualisation of spatial information using image data. A new model for imperfect geographic information is explored for this purpose: a twofold fuzzy region model; which is an extension of both the fuzzy region model and the egg/yolk model. This model is used to interpret spatial classifications and its imperfections in a new way. By defining different operators on the model an imprecise quality report can be developped for geographic databases that uses the imperfect spatial classification as reference information. The model makes it possible, despite the large amount of imperfection in spatial classifications, to use them for rather accurate error detection on geographic databases.

Evaluating and improving the accuracy of 3D models from UAS imagery

Abstract: Aerial images taken by an Unmanned Aircraft System (UAS) are used to generate large georeferenced overviews of a terrain. In addition, structure from motion algorithms have enabled the creation of three dimensional (3D) digital terrain models of this terrain [1]. As applications using these 3D models become more demanding, the accuracy requirements grow. In our work we evaluate the real world accuracy of state of the art 3D reconstruction methods and propose a system to improve the results.