A new video steganography approach that relies on establishing an N-dimension space encoding for DCT residual coefficients. To begin with, an appropriate 8 × 8 DCT or 16 × 16 DCT residual coefficient block is chosen for embedding. This selection is made based on the embedding strength of the Transform Block (TB), random numbers, and the values of the coefficients. Next, the upper-frequency coefficients are standardized into an N-dimension array and then mapped to points within the constructed N-dimension space. The mapped values of these points are computed according to pre-defined mapping rules. By comparing the values of meta-data (the data to be embedded) with the mapping values, all the neighboring points of the previous point in the N-dimension space are searched. Subsequently, the coefficients of the N-dimension array are adjusted to the neighboring points where the mapping value matches the metadata value, thus realizing video steganography. This method enables the embedding of multiple metadata bits. Usually, it only necessitates the modification of a single DCT coefficient, which helps to minimize the embedding error. In the H.265/HEVC environment, the successful extraction of information shows the high robustness of the video steganography algorithm. Prior to information embedding, STC is employed to pre - process the data and conceal it within the carrier information. Even when the carrier information is subject to a certain degree of interference, the correct information can still be retrieved. The experimental findings verify the advantages of the proposed video steganography approach.