A Hardware Friendly Haze Removal Method and Its Implementation

被引:0
|
作者
Li, Minjiang [1 ]
Cui, Mingxu [2 ]
Chi, Jun [1 ]
Zeng, Xiaoyang [1 ]
Jing, Minge [1 ]
Fan, Yibo [1 ]
机构
[1] Fudan Univ, State Key Lab ASIC & Syst, Shanghai, Peoples R China
[2] OmniVis Technol Inc, Shanghai, Peoples R China
基金
中国国家自然科学基金;
关键词
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The images captured in hazy weather are always affected by haze, hence image dehazing has become a hot field in image processing. However, the existing dehazing algorithms are hard to be implemented on hardware to speed up because of high computational complexity. In this paper, we propose a novel hardware friendly image dehazing method to improve the image quality and save the execution time simultaneously. First, it selects airlight from the local patch with the current densest haze to obtain estimated airlight in real time. Then, it adopts the TDM (Time Division Multiplexing) strategy when searching the optimal transmission to save nearly 86.5% of hardware resources. Finally, it uses the fixed-point guided filter to improve the precision of the final refined transmission. The experimental results indicate it outperforms other dehazing algorithms in terms of SSIM and CIEDE2000. The hardware architecture for our proposed method trades off dehazing accuracy against throughput, which can obtain high-quality restored images with a high-throughput of 166. 67Mpixels/s when processing 80frames/s FULL-HD 1080p at 500MHz frequency.
引用
收藏
页码:73 / 77
页数:5
相关论文
共 50 条
  • [1] Hardware Implementation of a Fast and Efficient Haze Removal Method
    Shiau, Yeu-Horng
    Yang, Hung-Yu
    Chen, Pei-Yin
    Chuang, Ya-Zhu
    [J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2013, 23 (08) : 1369 - 1374
  • [2] Hardware Implementation for Haze Removal With Adaptive Filtering
    Zhang, Bin
    Wei, Jing
    [J]. IEEE ACCESS, 2019, 7 : 142498 - 142506
  • [3] Hardware Implementation for Real-Time Haze Removal
    Zhang, Bin
    Zhao, Jizhong
    [J]. IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 2017, 25 (03) : 1188 - 1192
  • [4] VLSI Implementation for an Adaptive Haze Removal Method
    Kuo, Yao-Tsung
    Chen, Wei-Ting
    Chen, Pei-Yin
    Li, Cheng-Hsien
    [J]. IEEE ACCESS, 2019, 7 : 173977 - 173988
  • [5] Algorithm and Hardware Implementation of Image Haze Removal Combining with Sky Recognition
    Zhou, Xiaobo
    Fan, Xiaodong
    [J]. PROCEEDINGS OF THE 2017 INTERNATIONAL CONFERENCE ON ELECTRONIC INDUSTRY AND AUTOMATION (EIA 2017), 2017, 145 : 126 - 131
  • [6] Hardware-Friendly Approximation for Swish Activation and Its Implementation
    Choi K.
    Kim S.
    Kim J.
    Park I.
    [J]. IEEE Transactions on Circuits and Systems II: Express Briefs, 2024, 71 (10) : 1 - 1
  • [7] A Hardware-Friendly High-Precision CNN Pruning Method and Its FPGA Implementation
    Sui, Xuefu
    Lv, Qunbo
    Zhi, Liangjie
    Zhu, Baoyu
    Yang, Yuanbo
    Zhang, Yu
    Tan, Zheng
    [J]. SENSORS, 2023, 23 (02)
  • [8] A hardware-friendly logarithmic quantization method for CNNs and FPGA implementation
    Jiang, Tao
    Xing, Ligang
    Yu, Jinming
    Qian, Junchao
    [J]. JOURNAL OF REAL-TIME IMAGE PROCESSING, 2024, 21 (04)
  • [9] A Hardware-Friendly Low-Bit Power-of-Two Quantization Method for CNNs and Its FPGA Implementation
    Sui, Xuefu
    Lv, Qunbo
    Bai, Yang
    Zhu, Baoyu
    Zhi, Liangjie
    Yang, Yuanbo
    Tan, Zheng
    [J]. SENSORS, 2022, 22 (17)
  • [10] A Low-Complexity Noise Removal Technique and its Hardware Implementation
    Matsubara, Takeaki
    Moshnyaga, Vasily G.
    Hashimoto, Koji
    [J]. TENCON 2010: 2010 IEEE REGION 10 CONFERENCE, 2010, : 716 - 719