Photonics integrated interference imaging method is a far-field imaging method developed in recent years, which aims at high quality imaging and system flattening. It is expected to reduce the energy consumption, volume, and weight of the system can be reduced to 1/10~1/100 under the same resolution. However, the existing systems have sparse characteristics for high-frequency signal sampling. When inverse Fourier transform (IFT) is used to solve the object intensity distribution, Gibbs ringing artifact appears in the sharp edge of the restored observation target, thus affecting the image quality. In order to suppress artifacts, entropy prior is proposed and the entropy penalty characteristics are studied. The maximum entropy algorithm is designed by using entropy prior and combining with the characteristics of photonics integrated interference imaging. In order to verify the performance of the method, a multi-layer hierarchical aperture arrangement structure with better performance is used for simulation, and peak signal-to-noise ratio (PSNR), structural similarity coefficient (SSIM), and mean square error (MSE) are used as image quality evaluation methods. Simulation results show that the maximum entropy algorithm can eliminate the artifacts caused by high-frequency sparse sampling. For the images obviously affected by ringing, MSE and SSIM can be improved by more than 50%, and PSNR can be improved by more than 10%.