Abstract: With the development of stealth technology, the RCS of high-speed moving targets such as aircraft, the RCS of missile is getting smaller, which requires a long-time accumulation to improve detection capability. This paper proposed a small target long-time accumulation algorithm based on K-neighborhood correlation and Radon Fourier transform（RFT）. Firstly, a rough estimation of target motion parameters is processed by the K-neighborhood correlation. Subsequently, a high efficiency accumulation is executed by the RFT. Experimental results with simulated and measured data indicate that the proposed method provides an effective and efficient way for small target detection with the loss of SNR better than 0.5 dB.