Compared with the two-dimensional discrete cosine transform (2DDCT) in JPEG, the two-dimensional discrete wavelet transform (2DDWT) is the core transform of the image compression system in the JPEG2000 standard. In many systems that need to process images in real time, such as digital cameras, remote sensing and telemetry, satellite communications, multimedia communications, portable cameras, mobile communications, etc., chips are needed to implement the image encoding, decoding and compression process. Although many researchers have studied the wavelet transform for image processing, most of them only focus on algorithm research, and pay little attention to the complexity of the algorithm hardware implementation. There is also little research on the hardware implementation of the wavelet transform for image processing. This paper studies the wavelet transform algorithm for image processing and its hardware implementation. A fast algorithm of \"Combining the data-extension procedure into the lifting-based DWT core\" proposed in the literature [13] is carefully analyzed, and an algorithm suitable for hardware implementation of the 5/3 wavelet transform based on the lifting method is proposed. The algorithm is simulated and verified in MATLAB to prove its correctness. The hardware structure of the algorithm was designed, simulated in Simulink of MATLAB, and the structure was described and simulated in register transfer level (RTL) of VHDL language. It was successfully integrated into Altera\'s FPGA device for verification. Compared with the traditional wavelet transform boundary processing method, this algorithm does not need an additional boundary extension process because its boundary extension process is embedded in the wavelet transform module, which reduces the amount of memory read and write during the wavelet transform process, thereby reducing memory usage, reducing power consumption, and improving hardware utilization and computing speed. Compared with the algorithm proposed in the literature [13], this algorithm does not need to add additional hardware computing modules, and has the characteristic of not changing the regular structure of the original lifting wavelet algorithm during hardware implementation. This wavelet transform hardware chip implementation is not only suitable for JPEG2000 5/3 lossless wavelet transform, but can also be used in various other real-time image compression processing hardware systems.
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore