Application of CED and HED techniques for Shockwave Detection with High-speed Schlirden and Shadowgraph images

Abstract

Abstract Schlieren and shadowgraph images are valuable tools in aerospace engineering due to their user-friendly nature. They are commonly used to gain insights into flow patterns, particularly in the context of high-speed phenomena. Investigating dynamic shock wave structures, such as shock reflection patterns and interactions between shock waves and boundary layers, involves the use of high-speed Schlieren and shadowgraph systems. These techniques are vital for understanding complex aerodynamic phenomena. Typically, sequences of Schlieren and shadowgraph images are used for qualitative analysis. These image sequences serve as the foundation for understanding and interpreting the observed flow patterns. Dealing with extensive data from high-speed Schlieren and shadowgraph images often requires image processing methods. This includes techniques like background subtraction, edge recognition, and shock detection to enhance the quality of data and enable detailed analysis. The phenomenon of shock wave detection from a high-speed airflow over a forward-facing step is discussed in this article. This real-world example illustrates the practical application of Schlieren and shadowgraph imaging and the importance of accurate shockwave detection by comparing two different image processing methods for shockwave detection: Canny Edge Detection (CED) and Holistically-Nested Edge Detection (HED). According to the data, HED performs better than CED at correctly identifying shockwaves in Schlieren and shadowgraph images.