Processing of high-speed videos of shock wave boundary layer interactions
- PDF / 2,651,736 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 11 Downloads / 158 Views
ORIGINAL PAPER
Processing of high-speed videos of shock wave boundary layer interactions Samee Maharjan1
· Dag Bjerketvedt1 · Ola Marius Lysaker1
Received: 26 June 2019 / Revised: 16 March 2020 / Accepted: 7 September 2020 © The Author(s) 2020
Abstract This paper presents a framework for processing high-speed videos recorded during gas experiments in a shock tube. The main objective is to study boundary layer interactions of reflected shock waves in an automated way, based on image processing. The shock wave propagation was recorded at a frame rate of 500,000 frames per second with a Kirana high-speed camera. Each high-speed video consists of 180 frames, with image size [768 × 924] pixels. An image processing framework was designed to track the wave front in each image and thereby estimate: (a) the shock position; (b) position of triple point; and (c) shock angle. The estimated shock position and shock angle were then used as input for calculating the pressure exerted by the shock. To validate our results, the calculated pressure was compared with recordings from pressure transducers. With the proposed framework, we were able to identify and study shock wave properties that occurred within less than 300 µsec and to track evolveness over a distance of 100 mm. Our findings show that processing of high-speed videos can enrich, and give detailed insight, to the observations in the shock experiments. Keywords Image processing · Front tracking · Shock wave · High-speed videos
1 Introduction The introduction of charge-coupled device (CCD) and complementary metal-oxide semiconductor (CMOS) technology revolutionized high-speed photography during 1980s and 1990s. Today, there are numerous types of high-speed cameras that operate at a frame rate of more than a million images per second with resolution of one mega pixel. This development enables researchers to capture fast phenomena like the shock wave propagation and gas explosion in a video as a series of images [1]. Based on the advanced image and computer technology, these images now give an alternative source to estimate the shock wave characteristics like shock speed, pressure etc. [2,3]. However, in early days, images were mainly used for visualization of the phenomena [4]. From an image processing perspective, the ability to extract the desired information automatically was limited. Image processing is an interdisciplinary research field, where the aim is to extract some desired information from
B 1
Samee Maharjan [email protected] University of South-Eastern Norway, Kjølnes ring 56, 3918 Porsgrunn, Norway
images. It is widely used for object detection, object location, classification, segmentation and motion detection, among others. Based on improved computer power, sensor development and algorithmic progress, the applications have been broaden into various other fields, for example security system, road safety, document enhancing and gas dynamics. The study of a shock wave generally called shock on the other hand is one of the important research area in g
Data Loading...
