In traditional coded aperture imaging (CAI), the main problem with the deconvolution method is that the noise may dominate the decoding process. Focusing on this problem, a synthetic CAI (s-CAI) method is proposed with a potential to improve the signal-to-noise ratio (SNR) and spatial resolution of CAI. A series of raw coded images are recorded while the pinhole array is varied to different structures, and an iterative decoding algorithm is developed to reconstruct the object image by synthetically using all of these pinhole array structures and corresponding decoded images. Because the proposed iterative decoding algorithm has prominent advantage in noise immunization, and more spatial light components are acquired in comparison with common single-exposure CAI, both the image resolution and SNR are remarkably enhanced.