In the present study, optimization on β-glucanase-assisted extraction was made in order to isolate waxy barley starch from domestic cultivar using the D-optimal design suitable for response surface methodology (RSM). The results demonstrated that the amount of enzyme was found to be a major influencing factor on the extraction yield, which was substantially increased by increasing the amount of enzyme. It was also influenced by the reaction time and amount of water addition; however, the two factors were less influential than the amount of enzyme. The optimized condition by RSM for the reaction time was found to be 2.63 hours and amount of enzyme 1.7%, and amount of water addition 4.38 times the weight of raw material. With the enzyme treatment, the starch content in residues (R), particularly in R1 and R5, was reduced considerably, resulting in an increase in the extraction yield and therefore primarily and effectively releasing B-type starch small granule confirmed by scanning electronic microscopy. In addition, the study determined the physicochemical properties of isolated waxy starch (i.e., purity, water adsorption capacity, thermal properties, rheology and starch morphology) and compared them with those from the enzyme-not treated sample. It was found that they were almost similar to each other, except for the purity of starch, which was lower in the enzyme-treated sample than in the enzyme-not treated one.