Endoscopic ultrasonography (EUS) is considered the most sensitive imaging method for the diagnosis of chronic pancreatitis (CP). Several investigators have shown that EUS findings of CP correlate with the presence of CP on endoscopic retrograde pancreatography (ERP). In general, for diagnosing CP using EUS, the presence or absence of the following EUS criteria is determined: hyperechoic foci, hyperechoic strands, lobularity, shadowing calcifications, cysts, hyperechoic duct margins, visible side branches, main pancreatic duct dilatation, and main pancreatic duct irregularity. Using these criteria, we reviewed the number of EUS criteria required to diagnose early CP and whether each EUS criterion correlates with the severity of CP on ERP. CP is likely when more than three criteria (for "early CP") or more than five criteria (for "moderate CP") are present. Moreover, each EUS criterion has its own importance at each ERP classification level. However, the obtained images can be operator dependent, and interobserver variability may affect interpretation of CP by EUS. Therefore, we performed a quantitative computer analysis of parenchymal echogenicity and compared it with the EUS diagnosis of CP so that the diagnosis of CP on the basis of EUS criteria could be objectively supported by the quantitative analysis of EUS images. In conclusion, EUS can objectively distinguish between a normal pancreas and CP, and can be used to evaluate the severity of the CP. EUS is a useful modality for diagnosing CP and is relatively less invasive than other available modalities.