| An exposure time is set (step 11), and a position of a probe (2) is measured a plurality of times (step 12). A position measurement error is calculated in a plurality of patterns of average numbers of times (step 13). A correspondence table of measuring times and the positioning errors being calculated is calculated (step 14). A correspondence table being prepared of measuring times and hand shake errors is used to calculate a correspondence table of measuring times and (hand shake errors + the positioning errors) for the exposure time being set (step 15). For all exposure times being candidates to be set, steps 11 to 15 are repeated to calculate a correspondence table of measuring times and position measurement errors (step 16). One of the exposure times and one of the average numbers of times, at which a position measurement error reaches the minimum, are set (step 17). |
