Wyniki wyszukiwania - BazTech

1

Parametryzacja geometrii sceny dla estymacji głębi w dekoderze wizji wszechogarniającej

Szydełko Błażej, Klóska Dominika, Dziembowski Adrian

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2022

|

nr 4

388--391

PL

Technika MPEG Immersive Video (MIV) jest rozwijana w celu efektywnego dostarczania wizji w systemie swobodnej nawigacji. W technice tej przewidziano brak przesyłania informacji o geometrii sceny, a rekonstruowanie jej po stronie dekodera w czasochłonnym procesie estymacji map głębi. Sposobem usprawnienia procesu estymacji jest wykorzystanie dodatkowej informacji o geometrii sceny przesyłanej jako sparametryzowane cechy map głębi. W artykule skupiono się na analizie skuteczności istniejącego rozwiązania, rozszerzając je o rekurencyjną ekstrakcję cech.

EN

The MPEG Immersive Video (MIV) technique is being developed for efficient delivery of immersive video. In one scheme, instead of transmitting geometry information, the technique aims to reconstruct it at the decoder side in a time-consuming depth map estimation process. The way to improve the estimation process is to use additional information about the scene geometry transmitted as parametrized features of the depth maps. This paper focuses on the efficiency analysis of an existing solution, extending it with recursive feature extraction.

2

Applying the Tilt depth and Tilt Euler techniques of gravity data to decipher the basement depth in Sichuan Basin, China

Chen Qing, Dong Yunpeng, Tan Xianfeng, Wang Jiabei, Huang Xiaoyu, Chen Hao

Acta Geophysica

|

2021

|

Vol. 69, no. 6

2173--2186

EN

The tilt angle (i.e., TDR) provides an efficient way to recognize the horizontal locations of multi-source geological bodies at different depths and inclination angles. The tilt-depth method was initially derived by applying magnetic formulas and used to calculate the depth of magnetic sources. Recently researchers have attempted to extend this method to interpret depths in gravity field data. The tilt-depth method of gravity anomalies (i.e., GTilt-depth) could capture the depth of a buried source effectively, which makes it superior at deciphering the basement relief. Meanwhile, Tilt-Euler deconvolution (i.e., Euler deconvolution of TDR) has been utilized for estimating a source’s position from gridded data automatically, which requires no structural index. However, analytical singularities can be produced when performing inversion with the Tilt-Euler deconvolution owning to the derivatives of TDR being incalculable when the horizontal derivative is zero. The improved Tilt-Euler deconvolution provided an efficient way to eliminate analytical singularities and obtain more stable solutions. The results from the theoretical model show that the GTilt-depth method and improved Tilt-Euler deconvolution could be applied to calculate the buried depths more accurately and effectively. Application of these methods shows that they are able to capture more detailed features, and provide more straightforward and accurate results of depth, than traditional methods. Furthermore, the results obtained from the gravity data in Sichuan Basin show that the basement depth ranges from 3 to 11 km, and 3 to 7 km in the central uplift, which contains a local depression with a depth of 8 km. The basement exhibits a general pattern of “shallow in middle and deep in east and west”, which is consistent with the results revealed by gravityseismic jointly interpreted profile. This research provides a better indication of the basement structure when interpreting the regional geology in Sichuan Basin.

3

Realizacja eksperymentalnego systemu telewizji swobodnego punktu widzenia z łukowym ustawieniem kamer

Dziembowski A., Kuehn A., Łuczak A., Mieloch D., Wegner K.

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2014

|

nr 6

161--164

PL

Opisano eksperymentalny system telewizji swobodnego punktu widzenia zbudowany w Katedrze Telekomunikacji Multimedialnej i Mikroelektroniki Politechniki Poznańskiej. Przedstawiono szczegóły dotyczące budowy systemu wraz z metodami jego kalibracji, a także rezultaty w postaci przykładowej sekwencji wielowidokowej.

EN

This paper describes the experimental FTV system made at Chair of Multimedia Telecommunications and Microelectronics. We presented all details relating to the system including calibration methods, as well as the outcome - exemplary multiview sequence.

4

Interactions of partially conductive cracks with eddy currents in non-destructive evaluation

Janoušek L, Smetana M., Alman M.

Przegląd Elektrotechniczny

|

2011

|

R. 87, nr 5

59-61

EN

The paper analyses interactions of induced eddy currents with partially conductive cracks using numerical calculations. A cracked plate specimen is non-destructively inspected using a standard procedure. Uncertainty in the crack depth estimation from the gained eddy current response signals is predicted. Enhanced evaluation of real partially conductive cracks is proposed according to the presented results.

PL

W artykule omówiono numeryczną symulację prądów wirowych w obecności częściowo przewodzących pęknięć materiału. Wzorcową płytkę z pęknięciem poddano standardowej procedurze testującej, po czym dokonano estymacji głębokości pęknięcia na podstawie wzmocnionego sygnału generowanego przez prądy wirowe w obecności defektu. Na podstawie analizy wyników zaproponowano rozszerzoną metodę ewaluacji rzeczywistych defektów częściowo przewodzących.

5

Efficient Implementation of Stereovision Algorithms for Graphics Processing Unit in DirectShow Technology

Szajerman D., Pelczynski P.

Machine Graphics and Vision

|

2011

|

Vol. 20, No. 2

139-156

EN

Stereovision is a passive technique for estimation of depth in 3D scenes. Unfortunately, depth estimation in this imaging technique is computationally demanding. We show that stereovision matching algorithms can be efficiently mapped onto the present-day graphics processing units (GPUs). A number of modifications to the original image disparity estimation algorithm have been proposed that make running its computation on GPU platforms particularly efficient. A complete depth estimation system was implemented in GPU, covering correction of camera distortions, image rectification and disparity estimation. To obtain modularity of developed software, the DirectShow multimedia technology was used. Examples, computed depth maps are shown, and time performances of the proposed algorithms are outlined. The developed system has proved the usefulness of both GPU implementation and the DirectShow technology in scene depth estimation.

6

Computer vision based navigation systems —requirements and proposed system idea

Morański M.

Challenges of Modern Technology

|

2011

|

Vol. 2, no. 2

16--18

EN

This article highlights the main challenges and constrains concerning computer vision based navigation systems. A few special conditions in which the systems have to work are pointed out. Different approaches to sense the surrounding environment are reviewed. Finally, the navigation system which incorporates two computer vision techniques: time of flight and stereovision is proposed.