In this paper, a novel triple-image encryption and hiding algorithm is proposed by combining a 2D chaotic system, compressive sensing (CS) and the 3D discrete cosine transform (DCT). First, three grayscale plain images are sparsely represented by the 2D discrete wavelet transform (DWT). The resulting sparse matrices are scrambled twice through index sort scrambling and 3D zigzag scrambling. Then, the measurement matrix generated by the 2D chaotic system is used to compress the scrambled matrices. Finally, the compressed matrices are embedded into a color carrier image using the 3D DCT to obtain a visually meaningful cipher image. It is worth mentioning that most of the existing meaningful encryption algorithms encrypt and embed single images, which cannot meet the requirements of the era of big data. However, in our scheme, three grayscale images are simultaneously encrypted and embedded into a single color carrier image, which, when compared with existing algorithms, greatly improves the encryption efficiency. Moreover, simulation experiments and contrastive analyses show that the proposed visual encryption scheme has excellent visual security, robustness, decryption quality and running efficiency.