A method for dye laser wavelength correction applied for the measurement of OH radical with FAGE (Fluorescence Assay by Gas Expansion) is researched in this article. Sufficiently stable concentration of OH radical is produced with thermal dissociation of H2O by using an alumel filament and the fluorescence is excited with 282 nm laser in a low pressure cell. The fluorescence is detected with a photomultiplier and a high speed data acquisition card, while the laser light is monitored by a photodiode, and both signals are handled by a LabVIEW program for further analysis. The data acquisition card is triggered by a positive TTL pulse generated by a digital delay generator, which is triggered by a rising edge of a synchronized output pulse of the dye laser. The LabVIEW program is used to determine the location of the OH excited line according to the fluorescence intensity of OH radical when the frequency of the dye laser is scanned. By scanning dye laser wavelength range in 281.97~282.28 nm, excitation spectrum of OH radical is recorded. In order to optimize system parameters and achieve a high signal-to-noise ratio, the effects of the humidity, oxygen concentration, mass flow and pumping speed on fluorescence intensity and lifetime are studied at Q12 line and less than ±1.9% fluctuations of the fluorescence intensity is obtained. With analysis of the reaction mechanism of the thermal dissociation of H2O, it is concluded that reaction of oxygen and water is a major source of OH radical. Laser output wavelength is scanned in a small range around Q12 line to find out the exact exciting line and then correct the laser’s output, which might slightly shift due to the environmental change and leads to reduction of fluorescence intensity. The wavelength correction procedure is implemented many times and the results show that the systematic error of the instrument is less than 0.1 pm. According to the experimental results, this method meets the needs of quantitative accurate measuring tropospheric OH radical by FAGE.