A 4K-Capable FPGA Implementation of Single Image Haze Removal Using Hazy Particle Maps

Abstract

This paper presents a fast and compact hardware implementation using an efficient haze removal algorithm. The algorithm employs a modified hybrid median filter to estimate the hazy particle map, which is subsequently subtracted from the hazy image to recover the haze-free image. Adaptive tone remapping is also used to improve the narrow dynamic range due to haze removal. The computation error of the proposed hardware architecture is minimized compared with the floating-point algorithm. To ensure real-time hardware operation, the proposed architecture utilizes the modified hybrid median filter using the well-known Batcher’s parallel sort. Hardware verification confirmed that high-resolution video standards were processed in real time for haze removal.

Single Image Haze Removal from Image Enhancement Perspective for Real-Time Vision-Based Systems

“…More specifically, when compared to the design of Park et al [ 48 ], the utilization of registers, LUTs, and memory is reduced by , and , respectively; and the processing speed is increased by approximately thrice. Compared to the design of Ngo et al [ 17 ], the reduction rates in the utilized registers, LUTs, and memory are , and , respectively, and the processing speed is slightly improved. Thus, coupled with the algorithm performance evaluation in Section 4 , it can be concluded that our proposed algorithm is of paramount importance because of its superior performance and an efficient hardware prototype that is capable of handling high-quality video streams in real-time.…”

“…Accordingly, the maximum video resolution that the design of Park et al [ 48 ] can handle is solely super video graphics array (SVGA). Ngo et al [ 17 ] developed a 4K-capable IP of the algorithm proposed by Kim et al [ 15 ], and it consists of 70,864 registers, 56,664 LUTs and Mbits. The highest possible attainable processing speed is MHz, which is responsible for its 4K capability.…”

“…The DCP has also been developed in many directions to overcome its shortcomings of slow processing rate and unpredictable performance in sky regions [ 10 , 11 , 12 , 13 , 14 , 15 , 16 ]. A fast dehazing method proposed by Kim et al [ 15 ] and its corresponding 4K-capable intellectual property (IP) presented in Reference [ 17 ] are cases considered. By exploiting the modified hybrid median filter, which possesses excellent edge-preserving characteristics, to estimate the atmospheric veil, their method eliminated the need for soft-matting refinement, resulting in considerably faster processing speed, albeit a slight degradation in image quality.…”

Improved Color Attenuation Prior for Single-Image Haze Removal

“…Additionally, a sophisticated adaptive tone remapping algorithm [31] also is used, as opposed to the simple tone remapping method used by Tarel et al [16]. As a result, this algorithm is a significant improvement upon the fast haze removal using median filters, and its hardware design presented by Ngo et al [35] provides an efficient means of handling large images in real time. An issue arising from the assumption that the atmospheric light is pure white (A = 1) after local white balance is that noticeable background noise may occur in the dehazed image.…”

Single-Image Visibility Restoration: A Machine Learning Approach and Its 4K-Capable Hardware Accelerator

“…Furthermore, a novel hardware architecture was developed to realize the modified hybrid median filter. Although the previously developed architecture based on Batcher’s sorting network [ 18 ] is considerably compact and fast, the proposed design, which exploits both sorting and merging networks, is established to be even more efficient. This newly proposed architecture significantly contributes to the 4K capability of the proposed hardware accelerator.…”