针对储油罐底板易形成腐蚀、点蚀、穿孔等缺陷的问题，提出了一种基于地磁场的新型无损检测技术，以人工腐蚀缺陷的试板为研究对象，进行了弱磁检测。采用概率法和Symlets小波基分解法对检测结果进行分析，对比了两种方法在缺陷信号识别上的优缺点，最后通过在现场检测验证了弱磁检测方法的可靠性。

In this paper,a new type of nondestructive testing technology based on the geomagnetic field is proposed for testing defects which are easily formed in the oil tank bottom plate such as corrosion, pitting, perforation. The artificial corrosion defects on plate were used as the research object. Two data processing methods of probability and Symlets wavelet decomposition were adopted for the analysis of the testing results, their advantages and disadvantages on the defect signal recognition were compared, and finally the reliability of detection method was verified by the in-site testing.