The resolution of terahertz images based on the near-field synthetic aperture radar (NSAR) imaging mode is low due to the narrow antenna beamwidth and the electromagnetic wave loss that was ignored by previous algorithms. We propose a super-resolution terahertz NSAR imaging algorithm. There are two algorithm steps: First, we use the forward mathematical model of the NSAR imaging system and the back-projection algorithm (BPA) to calculate the point spread function (PSF). The next step is to deconvolve the NSAR image reconstructed by BPA to enhance its lateral resolution. To evaluate the effectiveness of our proposed method, we conducted both simulations and experiments. The experiment results show that the proposed method achieves a resolution of 0.67λ, which is a significant improvement. Overall, the proposed method has great potential to enhance the resolution of THz images and advance the use of THz technology in various fields.