To improve the lateral resolution and contrast and restrain the peak sidelobe energy of ultrasonic echo images, a sidelobe canceller algorithm applied to the ultrasonic imaging is proposed. The proposed method calculates the weight vector by eigenspace-based generalized sidelobe canceller algorithm. The coherence factor of the ultrasonic echo signal is used to optimize the obtained weight vector. Then, the common point target and dark speck are imaged. The simulation results show that the proposed method improves the lateral resolution by 60% compared with delay-and-sum algorithm, and has a improvement of 112% in contrast compared with the eigenspace algorithm. Experimental results based on an actual ultrasonic system agree well with the simulated results, which proves the effectiveness of the proposed method.