Browsing by Author "Karvonen, Juha"
Now showing 1 - 5 of 5
- Results Per Page
- Sort Options
- Barnsley Type Fractal Transforms for Image Compression and Recognition
Helsinki University of Technology | Licentiate thesis(1996) Karvonen, Juha - Compaction of C-band synthetic aperture radar based sea ice information for navigation in the Baltic Sea
Doctoral dissertation (article-based)(2006-12-08) Karvonen, JuhaIn this work operational sea ice synthetic aperture radar (SAR) data products were improved and developed. A SAR instrument is transmitting electromagnetic radiation at certain wavelengths and measures the radiation which is scattered back towards the instrument from the target, in our case sea and sea ice. The measured backscattering is converted to an image describing the target area through complex signal processing. The images, however, differ from optical images, i.e. photographs, and their visual interpretation is not straightforward. The main idea in this work has been to deliver the essential SAR-based sea ice information to end-users (typically on ships) in a compact and user-friendly format. The operational systems at Finnish Institute of Marine Research (FIMR) are currently based on the data received from a Canadian SAR-satellite, Radarsat-1. The operational sea ice classification, developed by the author with colleagues, has been further developed. One problem with the SAR data is typically that the backscattering varies depending on the incidence angle. The incidence angle is the angle in which the transmitted electromagnetic wave meets the target surface and it varies within each SAR image and between different SAR images depending on the measuring geometry. To improve this situation, an incidence angle correction algorithm to normalize the backscattering over the SAR incidence angle range for Baltic Sea ice has been developed as part of this work. The algorithm is based on SAR backscattering statistics over the Baltic Sea. To locate different sea ice areas in SAR images, a SAR segmentation algorithm based on pulse-coupled neural networks has been developed and tested. The parameters have been tuned suitable for the operational data in use at FIMR. The sea ice classification is based on this segmentation and the classification is segment-wise rather than pixel-wise. To improve SAR-based distinguishing between sea ice and open water an open water detection algorithm based on segmentation and local autocorrelation has been developed. Also ice type classification based on higher-order statistics and independent component analysis have been studied to get an improved SAR-based ice type classification. A compression algorithm for compressing sea ice SAR data for visual use has been developed. This algorithm is based on the wavelet decomposition, zero-tree structure and arithmetic coding. Also some properties of the human visual system were utilized. This algorithm was developed to produce smaller compressed SAR images, with a reasonable visual quality. The transmission of the compressed images to ships with low-speed data connections in reasonable time is then possible. One of the navigationally most important sea ice parameters is the ice thickness. SAR-based ice thickness estimation has been developed and evaluated as part of this work. This ice thickness estimation method uses the ice thickness history derived from digitized ice charts, made daily at the Finnish Ice Service, as its input, and updates this chart based on the novel SAR data. The result is an ice thickness chart representing the ice situation at the SAR acquisition time in higher resolution than in the manually made ice thickness charts. For the evaluation of the results a helicopter-borne ice thickness measuring instrument, based on electromagnetic induction and laser altimeter, was used. - Comparison of C- and L-band synthetic aperture radar images for sea ice motion estimation
Sähkötekniikan korkeakoulu | Master's thesis(2013-10-21) Lehtiranta, JonniPairs of consecutive C-band SAR images are routinely used for sea ice motion estimation. New Earth-observing satellites with SAR instruments of other wavelengths are being introduced. These seem promising for the sea ice regime, especially in the Baltic Sea that lacks multiyear ice and icebergs. In this work, SAR images acquired using different wavelengths are applied for sea ice motion estimation, and the suitability is compared. The work will also investigate motion estimation by using an image pair of different wavelengths. Using images of both C-band and L-band presents more opportunities for calculating motion within a shorter time frame, but presents challenges arising from the fundamentally different nature of reflection in the images. A motion estimation program was written for this purpose, and a method for calculating sea ice motion from consecutive SAR images, using two images from different frequency bands, is presented. The difficulties arising from differences in wavelength, resolution and reflection characteristics are investigated for sea ice in the Baltic Sea and an algorithm that works for C- and L-band and mixed image pairs is presented. - Reaaliaikainen massapuun kuorisuusastelajittelija
Helsinki University of Technology | Master's thesis(1991) Karvonen, Juha - Retinotopic Maps, Spatial Tuning, and Locations of Human Visual Areas in Surface Coordinates Characterized with Multifocal and Blocked fMRI Designs
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä(2012) Henriksson, Linda; Karvonen, Juha; Salminen-Vaparanta, Niina; Railo, Henry; Vanni, SimoThe localization of visual areas in the human cortex is typically based on mapping the retinotopic organization with functional magnetic resonance imaging (fMRI). The most common approach is to encode the response phase for a slowly moving visual stimulus and to present the result on an individual's reconstructed cortical surface. The main aims of this study were to develop complementary general linear model (GLM)-based retinotopic mapping methods and to characterize the inter-individual variability of the visual area positions on the cortical surface. We studied 15 subjects with two methods: a 24-region multifocal checkerboard stimulus and a blocked presentation of object stimuli at different visual field locations. The retinotopic maps were based on weighted averaging of the GLM parameter estimates for the stimulus regions. In addition to localizing visual areas, both methods could be used to localize multiple retinotopic regions-of-interest. The two methods yielded consistent retinotopic maps in the visual areas V1, V2, V3, hV4, and V3AB. In the higher-level areas IPS0, VO1, LO1, LO2, TO1, and TO2, retinotopy could only be mapped with the blocked stimulus presentation. The gradual widening of spatial tuning and an increase in the responses to stimuli in the ipsilateral visual field along the hierarchy of visual areas likely reflected the increase in the average receptive field size. Finally, after registration to Freesurfer's surface-based atlas of the human cerebral cortex, we calculated the mean and variability of the visual area positions in the spherical surface-based coordinate system and generated probability maps of the visual areas on the average cortical surface. The inter-individual variability in the area locations decreased when the midpoints were calculated along the spherical cortical surface compared with volumetric coordinates. These results can facilitate both analysis of individual functional anatomy and comparisons of visual cortex topology across studies.