Optimization of the GPU-based data evaluation for the low coherence interferometry

2018 ◽  
Vol 85 (11) ◽  
pp. 680-690
Author(s):  
Yinan Li ◽  
Markus Kästner ◽  
Eduard Reithmeier
Keyword(s):  
Performance Optimization ◽  
Measurement Data ◽  
Measurement Range ◽  
Performance Comparison ◽  
Data Evaluation ◽  
Processing Unit ◽  
Low Coherence ◽  
Optical Interferometers ◽  
Marquardt Algorithm ◽  
Evaluation Algorithm

Abstract Optical interferometers as non-contact measurement devices are very desirable for the measurement of surface roughness and topography. Compared to phase shifting interferometers (PSIs) with a limited measurement range and a scan step of maximum λ/4, the optical interferometers like low coherence interferometers (LCIs) evaluating the degree of fringe coherence allow a larger vertical measurement range. Their vertical measurement range is only limited by the scan length allowed by the linear piezo stage and the coherence length of the light source. To evaluate the obtained data for a large range, the common LCIs require much computation time. To overcome this drawback, we present an evaluation algorithm based on the Hilbert-Transform and curve fitting (Levenberg–Marquardt algorithm) using Compute Unified Device Architecture (CUDA) technology, which allows parallel and independent data evaluation on General Purpose Graphics Processing Unit (GPGPU). Firstly, the evaluation algorithm is implemented and tested on an in-house developed LCI, which is based on Michelson configurations. Furthermore, we focus on the performance optimization of the GPU-based program using the different approaches to further achieve efficient and accurate massive parallel computing. Finally, the performance comparison for evaluating measurement data using different approaches is discussed in this paper.

scholarly journals Long-term observations of tropospheric particle number size distributions and equivalent black carbon mass concentrations in the German Ultrafine Aerosol Network (GUAN)

2016 ◽  
Vol 8 (2) ◽  
pp. 355-382 ◽  
Author(s):  
Wolfram Birmili ◽  
Kay Weinhold ◽  
Fabian Rasch ◽  
André Sonntag ◽  
Jia Sun ◽  
...  
Keyword(s):  
Black Carbon ◽  
Particle Number ◽  
Measurement Data ◽  
Urban Traffic ◽  
Data Evaluation ◽  
Size Distributions ◽  
Evaluation Procedures ◽  
Carbon Mass ◽  
Ultrafine Aerosol ◽  
Mass Concentrations

Abstract. The German Ultrafine Aerosol Network (GUAN) is a cooperative atmospheric observation network, which aims at improving the scientific understanding of aerosol-related effects in the troposphere. The network addresses research questions dedicated to both climate- and health-related effects. GUAN's core activity has been the continuous collection of tropospheric particle number size distributions and black carbon mass concentrations at 17 observation sites in Germany. These sites cover various environmental settings including urban traffic, urban background, rural background, and Alpine mountains. In association with partner projects, GUAN has implemented a high degree of harmonisation of instrumentation, operating procedures, and data evaluation procedures. The quality of the measurement data is assured by laboratory intercomparisons as well as on-site comparisons with reference instruments. This paper describes the measurement sites, instrumentation, quality assurance, and data evaluation procedures in the network as well as the EBAS repository, where the data sets can be obtained (doi:10.5072/guan).

scholarly journals Time interval measurement module implemented in SoC FPGA device

2016 ◽  
Vol 62 (3) ◽  
pp. 237-246 ◽  
Author(s):  
Grzegorz Grzęda ◽  
Ryszard Szplet
Keyword(s):  
Mass Measurement ◽  
Measurement Data ◽  
Measurement Instrument ◽  
General Concept ◽  
Time Interval ◽  
Processing Unit ◽  
Test Results ◽  
Interval Measurement ◽  
Time Interval Measurement ◽  
On Chip

Abstract We presents the design and test results of a picosecond-precision time interval measurement module, integrated as a System-on-Chip in an FPGA device. Implementing a complete measurement instrument of a high precision in one chip with the processing unit gives an opportunity to cut down the size of the final product and to lower its cost. Such approach challenges the constructor with several design issues, like reduction of voltage noise, propagating through power lines common for the instrument and processing unit, or establishing buses efficient enough to transport mass measurement data. The general concept of the system, design hierarchy, detailed hardware and software solutions are presented in this article. Also, system test results are depicted with comparison to traditional ways of building a measurement instrument.

Detailed Analysis of Experimental Investigations on Boundary Layer Transition in Wake Disturbed Flow

2006 ◽  
Author(s):  
Andrea Cattanei ◽  
Pietro Zunino ◽  
Thomas Schro¨der ◽  
Bernd Stoffel ◽  
Berthold Matyschok
Keyword(s):  
Boundary Layer ◽  
Measurement Data ◽  
Boundary Layer Transition ◽  
Data Evaluation ◽  
Experimental Investigations ◽  
Layer Transition ◽  
Disturbed Flow ◽  
Time Space ◽  
Separation Bubbles ◽  
Hot Film

In the framework of a co-operation between the University of Genoa and the Darmstadt University of Technology measurement data of a former investigation at Darmstadt, comprising measurements with surface-mounted hot-film sensors on the boundary layer transition in wake disturbed flow, were transferred to Genoa, then re-evaluated and in great detail analyzed, much further than the original data evaluation. In these experimental investigations at Darmstadt, the boundary layer transition with and without transitional separation bubbles was studied on a circular cylinder in cross flow. The comparison of hot-wire traverses with the surface-mounted hot-film distributions clearly indicated that the surface-mounted hot-film technique is a very suitable measurement technique to obtain reliable information on transition and separation phenomena with both high spatial and temporal resolution. The new data evaluation techniques applied to these data at Genoa further enhanced the insight into the details of the boundary layer transition and separation process. The surface-mounted hot-film data were evaluated by means of time-space diagrams for the first three statistical moments (mean, RMS and skewness), with which the origin and the extent of unsteady separation bubbles clearly could be seen. The results obtained from these data analyses on the one hand yield a considerable enhancement of the understanding of the periodically unsteady boundary layer transition process and on the other hand they form the basis for the application of surface-mounted hot-film sensors in more complex flow situations like e.g. in cold flow multistage turbine or compressor test rigs or even in the hostile environment of real aero engine compressors or turbines.

Large optical aspheric mirror profile data evaluation method based on Levenberg-Marquardt algorithm

2016 ◽  
Vol 37 (4) ◽  
pp. 555-560 ◽  
Author(s):  
Jin Qiuyu ◽  
Ma Fei ◽  
Liu Entao ◽  
Wang Yun ◽  
Zhao Weiqian
Keyword(s):  
Evaluation Method ◽  
Data Evaluation ◽  
Profile Data ◽  
Marquardt Algorithm ◽  
Levenberg Marquardt ◽  
Aspheric Mirror

scholarly journals Real-Time Implementation and Performance Optimization of Local Derivative Pattern Algorithm on GPUs

2018 ◽  
Vol 8 (6) ◽  
pp. 5457
Author(s):  
Nisha Chandran ◽  
Durgaprasad Gangodkar ◽  
Ankush Mittal
Keyword(s):  
High Resolution ◽  
Image Retrieval ◽  
Performance Optimization ◽  
General Purpose ◽  
Content Based Image Retrieval ◽  
Maximum Speed ◽  
Processing Unit ◽  
Efficient Retrieval ◽  
Local Derivative ◽  
Local Derivative Pattern

<p><span>Pattern based texture descriptors are widely used in Content Based Image Retrieval (CBIR) for efficient retrieval of matching images. Local Derivative Pattern (LDP), a higher order local pattern operator, originally proposed for face recognition, encodes the distinctive spatial relationships contained in a local region of an image as the feature vector. LDP efficiently extracts finer details and provides efficient retrieval however, it was proposed for images of limited resolution. Over the period of time the development in the digital image sensors had paid way for capturing images at a very high resolution. LDP algorithm though very efficient in content-based image retrieval did not scale well when capturing features from such high-resolution images as it becomes computationally very expensive. This paper proposes how to efficiently extract parallelism from the LDP algorithm and strategies for optimally implementing it by exploiting some inherent General-Purpose Graphics Processing Unit (GPGPU) characteristics. By optimally configuring the GPGPU kernels, image retrieval was performed at a much faster rate. The LDP algorithm was ported on to Compute Unified Device Architecture (CUDA) supported GPGPU and a maximum speed up of around 240x was achieved as compared to its sequential counterpart.</span></p>

Sign in / Sign up

Export Citation Format

Share Document