We present a method for transferring images with high spatial resolution through highly scattering media. Rather than achieving high resolution (HR) using a larger transmission matrix (TM), which is difficult to implement in practice, we propose an image coding scheme based on compressed sensing, which is capable of recovering object information at low sampling rates, to enhance TM-based image reconstruction. The essence of the proposed scheme can be considered as an exchange of spatial and temporal resolution. By appending extra images, which are compressive encoded signals of the high-frequency information of the original object, following transmission of the low-resolution image, HR image of the original object can be reconstructed from the measured speckle signals. The feasibility of the proposed method is validated by optical experiments. The proposed method achieves a remarkable resolution improvement without relying on a complex optical system, thereby indicating promise for existing applications involving image transmission through scattering media.