Abstract
This paper proposes a novel algorithm to compute the 2-D discrete wavelet transform (DWT) of high-resolution (HR) images on low-cost visual sensor and Internet of Things (IoT) nodes. The main advantages of the proposed segmented modified fractional wavelet filter (SMFrWF) are reduced computation (time) complexity and energy consumption compared to the state-of-the-art low-memory 2-D DWT computation methods. In particular, the conventional convolution-based DWT is very fast but requires large random access memory (RAM), as the entire image needs to be in the system memory. The fractional wavelet filter (FrWF) requires only a small RAM but has high complexity due to multiple readings of image lines. The proposed SMFrWF avoids the multiple readings of image lines, thus reducing the memory read access time and, thereby, the complexity. We evaluated the proposed SMFrWF through MATLAB simulations with 70 popular gray-scale test images of dimensions ranging from $256 \times 256$ up to $8192 \times 8192$ pixels. The results show that for images of size $2048\times 2048$ pixels, the proposed SMFrWF (with four segments per line) has 16.8% and 53.6% lower time complexities than the conventional DWT and FrWF, respectively. The proposed SMFrWF has also been modeled in a hardware description language (HDL) and implemented on an Artix-7 field-programmable gate array (FPGA) platform to evaluate the hardware performance. We observed that the proposed SMFrWF has 65% lower energy consumption than the FrWF (both implemented on the same board). Thus, the proposed SMFrWF appears suitable for computing the wavelet transform coefficients of HR images on low-cost visual sensors and IoT platforms.
Highlights
We demonstrate the compatibility of the proposed Modified Fractional Wavelet Filter (MFrWF) and segmented MFrWF (SMFrWF) with state-of-theart wavelet based image coding algorithms and evaluate the coding efficiency in terms of the peak-signal-to-noise-ratio (PSNR)
We have verified through extensive simulations and hardware implementations that the proposed MFrWF has substantially lower complexity than the conventional discrete wavelet transform (DWT) and fractional wavelet filter (FrWF), while generating exactly the same wavelet transform coefficients
We observed a trade-off between the memory requirement and computational complexity that is controlled by the number of line segments, which is a useful feature of the SMFrWF
Summary
The memory requirement of all these stripe-based approaches for a typical 512 × 512 gray-scale image is about 26 kB [20], which is more than the available RAM of many low-cost visual sensor nodes. The architecture proposed in [47] needs no temporal memory, it requires a large line buffer and a complex control scheme. The memory requirements of other state-of-the-art DWT architectures, such as [51], [52] are more than the RAM available on most of the low-cost sensor nodes. To the best of our knowledge, the proposed SMFrWF is the first attempt to develop an alternative to the FrWF that reduces the computation complexity for HR image wavelet transforms on low-memory portable multimedia devices.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have