超声电视成像测井中图像异样点修复处理改进方法研究

针对超声成像测井中因各种不确定因素而导致测井图像中出现异样点的现象，通过分析目前清除异样点所采取的常规方式的不足，提出一种能够有效解决此类问题的全变分修复算法；该方法先于选定区域内依据阈值大小识别待修复的异样点，随后采用该修复算法对此进行修复；并分别将其应用于标准测井图像与实际测井资料处理中并进行效果比对；实践证明：该算法能够完全恢复其图像的原始特征，为后期图像资料中此类状况的修复处理提供了有效的解决方案．

Improved approach to repairing image outliers in ultrasonic TV imaging logging

Aiming at the phenomenon that some outliers will appear in a logging image due to various uncertain fac- tors in the process of ultrasonic imaging logging, by analyzing the insufficiency of the normal method taken for re- moving the outliers, the authors proposes a total-variation repair algorithm that may effectively solve such problems. The method first identifies an outlier to be repaired according to the threshold within the selected scope, then ap- plies the algorithm to repair it. Then this repair algorithm is applied to process the standard logging image and the actual logging image, and then a comparison is made between the two results. It has been verified that the algorithm is capable of restoring the original features of the images completely and it provides an effective solution for the re- pair of image data in later step in such circumstances.