Blue horizontal-branch (BHB) stars are crucial for studying the structure of the Galactic halo. Accurate atmospheric parameters of BHB stars are essential for investigating the formation and evolution of the Galaxy. In this work, a data-driven technique named Stellar Label Machine (SLAM) is used to estimate the atmospheric parameters of Large Sky Area Multi-Object Fiber Spectroscopic Telescope low-resolution spectra for BHB stars with a set of A-type theoretical spectra as the training data set. We add color indexes ((BP-G), (G-RP), (BP-RP), (J-H)) during the training process to constrain the stellar temperature further. Finally, we derive the atmospheric parameters (Teff, logg, [Fe/H]) for 5355 BHB stars. Compared to existing literature results, our results are more robust, and after taking the color index into account the resulting precision of Teff and logg is significantly improved, especially for the spectrum with a low signal-to-noise ratio (S/N). Based on the duplicate observations with an S/N difference <20%, the random errors are around 30K, 0.1dex, and 0.12dex for Teff, logg, and [Fe/H], respectively. The stellar labels provided by SLAM are also compared to those from the high-resolution spectra in the literature. The standard deviation between the predicted star labels and the published values from the high-resolution spectra is adopted as the statistical uncertainty of our results. They are {sigma}(Teff)=76K, {sigma}(logg)=0.04dex, and {sigma}([Fe/H])=0.09dex, respectively.