Scholarlink

<p>To address the difficulty of registration for dynamic fluid image, a novel registration method based on temporal variation is proposed. During the experiment, we take candle flame and gun flame as study objects. Then the visible and infrared images are acquired by different cameras with different field angles. The similar point pairs of grayscale changes in ten consecutive frames are extracted, which serve as reference points to align these ten frames by using an affine transformation model. Finally, the image registration results are evaluated by subjective evaluation and objective parameter. After collecting actual flame data through the construction of a spectral path experiment, the registration of ten frames of images was completed. The root mean square error reduction rates of the registration results for candle flames and spray flames were 11.37% and 26.33%, respectively. The results show that the time-series variation method is effective for dynamic flame fluid image registration.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Dynamic fluid image registration method based on time-series variation

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Penghao Liao,
  • Fujiang Zeng,
  • Xiaoyan Wu,
  • Xiaomin Yang,
  • Rongzhu Zhang

摘要

To address the difficulty of registration for dynamic fluid image, a novel registration method based on temporal variation is proposed. During the experiment, we take candle flame and gun flame as study objects. Then the visible and infrared images are acquired by different cameras with different field angles. The similar point pairs of grayscale changes in ten consecutive frames are extracted, which serve as reference points to align these ten frames by using an affine transformation model. Finally, the image registration results are evaluated by subjective evaluation and objective parameter. After collecting actual flame data through the construction of a spectral path experiment, the registration of ten frames of images was completed. The root mean square error reduction rates of the registration results for candle flames and spray flames were 11.37% and 26.33%, respectively. The results show that the time-series variation method is effective for dynamic flame fluid image registration.