Study on evaluation methods for warship infrared simulation images
With the high efficiency, reusability, and low system risk of infrared scene simulation technology being widely applied in practical applications, researchers have put forward higher requirements for the authenticity of their simulation results. Therefore, how to reasonably and effectively evaluate the simulation results of infrared scenes has become a key issue for the further development of infrared scene simulation technology. However, there is currently no unified standard for infrared scene simulation evaluation in China, coupled with the complexity and diversity of simulation problems, as well as the lack of effective verification methods and means, the development of simulation evaluation research is relatively slow. This article focuses on studying the similarity between ship infrared simulation images and real infrared images, and quantitatively evaluates them. Starting from typical features such as texture, contour, and size of the images, several feature extraction algorithms are used for image quality evaluation. The similarity evaluation of infrared simulation images in the 3-5μm band at different times is conducted, and the effectiveness and feasibility of the three evaluation methods are verified through experiments, providing a new reference scheme for the field of infrared simulation evaluation
Authors
- Wang Lv (ORCID: https://orcid.org/0009-0007-2014-0480)
- Kefeng Sun (ORCID: https://orcid.org/0009-0002-3580-6365)
- Hongliang Cheng (ORCID: https://orcid.org/0009-0004-9994-7823)
- Yuan Liang (ORCID: https://orcid.org/0000-0003-2330-1837)
- Meng Gao (ORCID: https://orcid.org/0000-0002-5818-1639)
- Man Wang (ORCID: https://orcid.org/0009-0004-0891-7259)
- Xicheng Wang (ORCID: https://orcid.org/0000-0002-2719-2538)
- Bing Yu (ORCID: https://orcid.org/0000-0003-3220-2865)
- Jianghui Wu
- Hemin Meng
Publication Details
- Published
- 2024-12-04
- DOI
- https://doi.org/10.1117/12.3036873
- Primary Topic
- Infrared Target Detection Methodologies
- Type
- article
- Field-Weighted Citation Impact
- 0.00