<p>With the rapid development of the Internet and communication technology, people are using video to transmit information more and more frequently on the network. Due to the large amount of information carried by video, which may also contain important contents such as personal privacy and business secrets, it may become the preferred target of illegal attackers. Therefore, this paper designs a video encryption algorithm based on a novel spatiotemporal chaotic system and an IFS fractal construction system. At the system level, this paper designs a new ICMIC Improved Arnold Coupled Chebyshev Map Lattice (IIACCML) system on top of Coupled Map Lattice (CML) spatiotemporal chaotic system. The results of dynamical behavior tests show that the IIACCML system has a larger parameter space as well as better chaotic properties than the traditional spatiotemporal chaotic system. At the algorithmic level, this paper adopts the reduced-dimensional frame-by-frame encryption strategy, and the encryption of each frame in the video is divided into block-level encryption and pixel-level encryption. The algorithm first utilizes the chaotic sequences generated by the IIACCML system as well as the Iterated Function System (IFS) fractal construction system and the fractal images to construct the indexed control sequences of the two rounds of encrypted disorder and diffusion links, and then utilizes the principle of indexed image conversion to realize multi-frame image encryption. Among them, the indexed image transformation principle can convert the color video encryption problem into grayscale video encryption problem by dimensionality reduction, which greatly improves the encryption efficiency. Simulation results and performance tests show that the algorithm has high security and encryption efficiency.</p>