Fundamental frequency extraction of acoustic signal is a key step for determination of elastic modulus of wood by acoustic-ultrasonic technology. Calculation of elastic modulus by fundamental frequency extracted by the existing fast Fourier transform (FFT) method is about 20% bigger than one which acquired by the standard mechanical method for noise and sensor resonant frequency and other factors. For this, the largest common divisor algorithm was proposed for extraction of fundamental frequency to construct an improved method of measuring elastic modulus of wood, and gave the detailed process of extracting the fundamental frequency. The poplar core samples were tested respectively by application the method proposed in the paper and FFT method. The calculated values of elastic modulus(Eu), as determined by the method proposed in the paper, and the calculated values of elastic modulus(Ef), as determined by FFT method, were in the range of 8.23~40.32Gpa and 7.94~51.87Gpa. Comparing with measured value of static modulus of elasticity (Es), ranged from 6.72Gpa to 36.35Gpa, as determined by the standard method of mechanics, error was decreased by 10% or so. Linear correlations were observed respectively, with a coefficient of determination of 0.94 between Eu and Es, and with a coefficient of determination of 0.86 between Ef and Es. Results showed significant correlations. The measured data show that compared with elastic modulus acquired by mechanical method, testing values are more consistent and better correlations by the method proposed in the paper than one by FFT method.