光学图像中空间物体条痕检测方法综述OA北大核心CSTPCD
Review of Space Object Streak Detection Methods in Optical Images
空间物体在光学图像中常常表现为拖长的条纹形状,即条痕,条痕的准确检测对空间物体的识别与测量具有重要意义.近年来,许多研究人员对空间物体条痕的检测方法进行研究,取得了丰硕的成果.随着空间物体探测能力的提高,暗弱条痕的检测在空间物体精确探测中变得越发重要.对当前光学图像中空间物体条痕检测方法的主要原理和步骤进行了综述,根据不同方法的特点进行了分类和总结,并选择了四种不同类别的代表性单帧图像空间物体条痕检测算法,利用模拟图像和实测图像对这四种方法进行了实验对比,根据实验结果分析了各方法的优缺点,最后对光学图像中的条痕检测方法的未来发展方向进行了展望.
Space objects frequently appear as elongated stripe patterns,known as streaks,in optical images.The accurate detection of streaks is of great significance for the recogni-tion and measurement of space objects.In recent years,many researchers have studied the detection methods of space object streaks and achieved fruitful results.With the improve-ment of the detection ability of space objects,the detection of faint streaks has become more and more important in realizing comprehensive and accurate space object detection.This paper provides an overview of the main principles and steps of space object streak detec-tion methods in optical images.It classifies and summarizes these methods based on their characteristics,and selects four representative single-frame image-based algorithms for space object streak detection from different categories.Experimental comparisons of these four methods are conducted using both simulated and measured images.Based on the experi-mental results,the advantages and disadvantages of each method are analyzed.Finally,the future development directions of streak detection methods in optical images are prospected.
王磊;张晓明;王建峰;曾显群;谢海闻;姜晓军
中国科学院 国家天文台光学天文重点实验室,北京 100101||中国科学院大学天文与空间科学学院,北京 100049中国科学院 国家天文台光学天文重点实验室,北京 100101中国科学院大学天文与空间科学学院,北京 100049||中国科学院 国家天文台长春人造卫星观测站,长春 130117
天文学
图像处理空间物体光学图像条痕检测
image processingspace objectoptical imagestreak detection
《天文学进展》 2024 (003)
473-493 / 21
科技创新2030"新一代人工智能"重大项目(2022ZD0117401);国家自然科学基金(12273063)
评论