图像处理技术在位场数据边界识别中的应用进展

Application progress of image processing techniques on potential-field edge detection

边界识别方法是重磁位场数据解释中非常重要的反演技术,是进一步深化解释工作的基础。在总结传统边界识别方法的基础上,阐述了图像处理技术在边界增强与边界拾取方面的应用现状,讨论了图像处理技术的优势与不足,最终指出利用人工智能深度学习技术中的图像分割方法,可提升边界识别的效果与自适应能力,为位场解释技术的自动化与智能化发展提供新的研究思路。

Edge detection is a very important inversion technique in the interpretation of gravity and magnetic potential-field data.It is helpful for further professional interpretation.Based on summarizing the traditional methods of edge detection,this paper describes the application situation of image processing techniques on edge enhancement and picking,and discusses their strengths and weaknesses.We also propose that the image segmentation in deep learning of artificial intelligence could improve the effect and adaptive ability of edge detection,and provide new research ideas for the automation and intelligent development of potential-field interpretation technology.