Frank Tintrup

Compression algorithms for classification of remotely sensed images

The paper presents a comparison of the principal lossy compression algorithms, vector quantization (VQ), JPEG and wavelets (WV) posterior KLT applied to multispectral remotely sensed images and evaluated by the classification algorithm K-NN (K-nearest neighbor). The main goal of the compression of remotely sensed images is a reduction of the huge requirements for downlink and storage. The Karhunen Loeve transform first removes the interband correlation to produce the principal components of the image which are then compressed by the principal algorithms. The quality evaluation was done by a supervised classification with the well known algorithm K-NN for remote sensing applications and the MSE for visual aspects. The obtained results of these accurate and particular analysis of the current compression techniques are quite surprisingly compared to other previous works.

Automatic land classification vs. data compression: a comparative evaluation

Presents an accurate comparison of three known lossy compression techniques. Already well established is the vector quantization and JPEG while the coding of transformed wavelet coefficients is a more recent technique. Multispectral remotely sensed images (Thematic Mapper) have been transformed by the Karhunen Loeve transform (KLT) before compression by the algorithms. As the goal of this accurate analysis is the compression for automatic classification of the images, the efficiency and quality of the techniques was evaluated by a supervised classification of the decoded images, the well known algorithm K-NN (k-nearest-neighbor) for remote sensing applications while the MSE (mean square error) was compared for visual aspects. The main goal of the compression of remotely sensed images is a reduction of the huge requirements for downlink and storage which are needed nowadays to allocate the very high and increasing quantity of data from recent sensors. The Karhunen Loeve transform first removes the interband correlation to produce the principal components of the images which are then compressed by the three principal algorithms. The obtained results of these particular and accurate analysis of the current compression techniques are quite surprisingly compared to other recent works where most compression techniques perform well for visual aspects (e.g. browsing).

Origin of motion in the human ureter: mechanics, energetics and kinetics of the myosin molecular motors

Background Ureteral peristalsis is the result of coordinated mechanical motor performance of longitudinal and circular smooth muscle layer of the ureter wall. The main aim of this study was to characterize in smooth muscle of proximal segments of human ureter, the mechanical properties at level of muscle tissue and at level of myosin molecular motors. Methods Ureteral samples were collected from 15 patients, who underwent nephrectomy for renal cancer. Smooth muscle strips longitudinally and circularly oriented from proximal segments of human ureter were used for the in vitro experiments. Mechanical indices including the maximum unloaded shortening velocity (Vmax), and the maximum isometric tension (P0) normalized per cross-sectional area, were determined in vitro determined in electrically evoked contractions of longitudinal and circular smooth muscle strips. Myosin cross-bridge (CB) number per mm2 (Ψ) the elementary force per single CB (Ψ) and kinetic parameters were calculated in muscle strips, using Huxleys equations adapted to nonsarcomeric muscles. Results Longitudinal smooth muscle strips exhibited a significantly (p<0.05) faster Vmax (63%) and a higher P0 (40%), if compared to circular strips. Moreover, longitudinal muscle strips showed a significantly higher unitary force (Ψ) per CB. However, no significant differences were observed in CB number, the attachment (f1) and the detachment (g2) rate constants between longitudinal and circular muscle strips. Conclusions The main result obtained in the present work documents that the mechanical, energetic and unitary forces per CB of longitudinal layer of proximal ureter are better compared to the circular one; these preliminary findings suggested, unlike intestinal smooth muscle, a major role of longitudinal smooth muscle layer in the ureter peristalsis.

Classification of Compressed Multispectral Data

The requirements of increasing resolution in remote sensing are leading to excessively large data volumes. Powerful compression techniques are therefore needed to reduce bandwidth needed for satellite downlinks and to ease storage requirements. In this chapter, we report on a comparison of the principal lossy compression algorithms. The analyzed techniques are Vector Quantization (VQ),JPEG and Wavelets (WV), which have been applied to multispectral remotely sensed images. Prior to compression by the different algorithms the Karhunen Loeve Transform has been used to remove the interband correlation and produce the principal components of the image. A supervised classification with the well known K-Nearest Neighbour algorithm for remote sensing applications has been used for quality evaluation while the Mean Square Error (MSE) was calculated in the second order.