[DCT-FPGA] verilog programming implementation, FPGA-based DCT transformation implementation

1. Software version

ISE14.7

2. Theoretical knowledge of this algorithm

       In order to effectively calculate DCT, many algorithms have been proposed. Generally, there are two kinds of DCT fast algorithms: indirect algorithm and direct algorithm. Both algorithms focus on butterfly structure, and the purpose is to reduce the calculation amount of multiplication and addition. . The indirect algorithm uses the relationship between DCT and DFT, DHT and other orthogonal transformations, and uses the DFT or DHT fast algorithm to calculate the DCT. The process of the indirect algorithm is simple, and the main job is to deal with the conversion between algorithms, so it is often necessary to add some additional operation steps. Since the application of other fast algorithms of transformation in DCT always has its own limitations, there are few algorithms at present. Some people use an indirect algorithm to calculate the DCT. The direct algorithm includes DCT transformation matrix decomposition and recursive algorithm. The difference is that the matrix decomposition uses the sparse matrix decomposition method to decompose the transformation matrix, while the recursive algorithm recursively generates a higher-order DCT matrix from a lower-order DCT matrix. It is said that the recursive algorithm is the inverse algorithm of the decomposition algorithm, but the recursive algorithm has better numerical stability than the matrix decomposition algorithm. 

       Two-dimensional DCT converts motion compensation errors or original picture information blocks into coefficient sets representing different frequency components, which has two advantages: First, video signals often concentrate most of their energy in a small range in the frequency domain. , so that only a small number of bits are required to describe the unimportant componentsÿ

Guess you like

Origin blog.csdn.net/ccsss22/article/details/124306752