In this work, we further developed the planar temperature measurement method based on thermally assisted laser-induced fluorescence (TALF) with a single femtosecond (fs) laser and investigated the temperature distribution in the nitrogen-diluted oxygen–hydrogen cross-jet non-premixed flames at 1 kHz. Fs-TALF thermometry was firstly calibrated and assessed in a series of unconfined methane/air laminar flames at atmospheric pressure. In the calibration process, although the calibrated energy difference between OH v′ = 0 and 1 energy level was different from the theoretical value, the measured fluorescence ratio accurately reflected the temperature change from 1600 K to 2200 K in different methane flames. By comparing the calculated OH fluorescence spectrum, we believed that the main reason was the overlap of the OH (1-1) and (0-0) fluorescence bands. After the calibration, experiments were conducted with changing hydrogen flow velocity from 60 to 100 m/s and constant oxidizer flow velocity, where the dilution ratio, D varies from 0.46 to 0.7, and the equivalence ratio, ϕ, varies from 0.5 to 0.9. It is found, within the calibration range, fs-TALF thermometry can correctly measure the flame temperature. However, with ϕ further increases, the difference between the measured and theoretical temperatures gradually increases. At the same time, it was noted that the error in the measured temperature rises gradually as D decreases. In conclusion, fs-TALF thermometry can measure turbulent non-premixed hydrogen flames accurately in a range of temperature with acceptable temporal and partial resolution, while the errors within 200 K and spatial resolution is 170μm. The main error source should be the collisional effect due to the different local concentration of N2 and H2O. The results of this work will contribute to the further development the temperature measurement method hydrogen turbulent flames.