基于多通道视觉注意力的模糊图像质量复原算法设计

doi:10.13229/j.cnki.jdxbgxb.20220684

摘要/Abstract

摘要：

为了提高摄像机成像质量和图像清晰度，提出了一种基于多通道视觉注意力的模糊图像质量复原算法。在多通道视觉注意力机制的基础上构建了编码网络，将模糊图像输入到编码网络中，对图像展开复原处理；同时，在复原处理过程中引入振铃检测，消除复原后图像中存在的振铃区域，提高复原图像的质量和清晰度。实验结果表明，本文算法的去振铃现象效果更好、图像复原后的质量更高、应用效率更理想。

关键词: 多通道视觉注意力, 模糊图像, 振铃效应, 编码网络, 图像质量复原

Abstract:

In order to improve the imaging quality of the camera and improve the image definition， a fuzzy image quality restoration algorithm based on multi-channel visual attention was proposed. Based on the multi-channel visual attention mechanism， the coding network was constructed. The blurred image is input into the coding network， and the image was restored. At the same time， the ringing detection was introduced in the restoration process to eliminate the ringing area in the restored image and improve the quality and clarity of the restored image. The experimental results show that the proposed algorithm has better effect of de ringing phenomenon， high quality after image restoration， and better application efficiency.

Key words: multi-channel visual attention, blurred image, ringing effect, coding network, image quality restoration

中图分类号:

TP391.41

陈烽,王浩. 基于多通道视觉注意力的模糊图像质量复原算法设计[J]. 吉林大学学报(工学版), 2023, 53(9): 2626-2631.

Feng CHEN,Hao WANG. Design of fuzzy image quality restoration algorithm based on multi⁃channel visual attention[J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(9): 2626-2631.

图/表 6

图1

图2

图3

图4

表1

图5

参考文献 18

1	黄彦宁, 李伟红, 崔金凯, 等. 强边缘提取网络用于非均匀运动模糊图像盲复原[J]. 自动化学报, 2021, 47(11): 2637-2653.
	Huang Yan-ning, Li Wei-hong, Cui Jin-kai, et al. Strong edge extraction network for non-uniform blind motion image deblurring[J]. Acta Automatica Sinica, 2021, 47(11): 2637-2653.
2	叶晓杰, 崔光茫, 赵巨峰, 等. 基于闪动快门的互补序列对的运动模糊图像复原[J]. 光子学报, 2020, 49(8): 161-175.
	Ye Xiao-jie, Cui Guang-mang, Zhao Ju-feng, et al. Motion blurred image restoration based on complementary sequence pair using fluttering shutter imaging[J]. Acta Photonica Sinica, 2020, 49(8): 161-175.
3	张玉波, 王建阳, 韩爽, 等. 一种非对称的轻量级图像盲去模糊网络[J]. 吉林大学学报: 理学版, 2023, 61(2): 362-370.
	Zhang Yu-bo, Wang Jian-yang, Han Shuang, et al. An asymmetric lightweight image blind deblurring network[J]. Journal of Jilin University (Science Edition), 2023, 61(2): 362-370.
4	王若状, 臧景峰, 张朋朋. 基于模拟曝光融合的非均匀光照图像增强[J]. 吉林大学学报: 理学版, 2023, 61(3): 601-611.
	Wang Ruo-zhuang, Zang Jing-feng, Zhang Peng-peng. Non-uniform illumination image enhancement based on simulated exposure fusion[J]. Journal of Jilin University (Science Edition), 2023, 61(3): 601-611.
5	钱旭淼, 段锦, 刘举, 等. 基于注意力特征融合的图像去雾算法[J]. 吉林大学学报: 理学版, 2023, 61(3): 567-576.
	Qian Xu-miao, Duan Jin, Liu Ju, et al. Image dehazing algorithm based on attention feature fusion[J]. Journal of Jilin University (Science Edition), 2023, 61(3): 567-576.
6	杨燕, 王志伟. 基于补偿透射率和自适应雾浓度系数的图像复原算法[J]. 通信学报, 2020, 41(1): 66-75.
	Yang Yan, Wang Zhi-wei. Image restoration algorithm based on compensated transmission and adaptive haze concentration coefficient[J]. Journal on Communications, 2020, 41(1): 66-75.
7	李澜, 巩彩兰, 黄华文, 等. 基于离散余弦变换的无人机耀斑图像恢复算法[J]. 光学学报, 2020, 40(19): 186-191.
	Li Lan, Gong Cai-lan, Huang Hua-wen, et al. Unmanned aerial vehicle glint image restoration algorithm based on discrete cosine transform[J]. Acta Optica Sinica, 2020, 40(19): 186-191.
8	范伟强, 刘毅. 基于自适应小波变换的煤矿降质图像模糊增强算法[J]. 煤炭学报, 2020, 45(12): 4248-4260.
	Fan Wei-qiang, Liu Yi. Fuzzy enhancement algorithm of coal mine degradation image based on adaptive wavelet transform[J]. Journal of China Coal Society, 2020, 45(12): 4248-4260.
9	李荣华, 唐智超, 朴俊峰, 等. 偏振参数最优重构的水下降质图像清晰化方法[J]. 红外与激光工程, 2021, 50(6): 305-313.
	Li Rong-hua, Tang Zhi-chao, Pu Jun-feng, et al. Underwater degraded image-sharpening method based on optimal polarization parameter reconstruction[J]. Infrared and Laser Engineering, 2021, 50(6): 305-313.
10	杨晶显, 张帅, 刘继春, 等. 基于VMD和双重注意力机制LSTM的短期光伏功率预测[J].电力系统自动化, 2021, 45(3): 174-182.
	Yang Jing-xian, Zhang Shuai, Liu Ji-chun, et al. Short-term photovoltaic power prediction based on variational mode decomposition and long shortterm memory with dual-stage attention mechanism[J]. Automation of Electric Power Systems, 2021, 45(3): 174-182.
11	张艺璇, 郭斌, 刘佳琪, 等. 基于多级注意力机制网络的APP流行度预测[J]. 计算机研究与发展, 2020, 57(5): 984-995.
	Zhang Yi-xuan, Guo Bin, Liu Jia-qi, et al. APP popularity prediction with multi-level attention networks[J]. Journal of Computer Research and Development, 2020, 57(5): 984-995.
12	姚钦舟, 庄苏锋, 屠大维, 等. 水下透视投影图像非线性畸变校正方法[J]. 仪器仪表学报, 2020, 41(2): 176-183.
	Yao Qin-zhou, Zhuang Su-feng, Tu Da-wei, et al. Non-linear distortion correction method for underwater perspective projection image[J]. Chinese Journal of Scientific Instrument, 2020, 41(2): 176-183.
13	黄超, 郭浩, 高岩, 等. 基于栈式自编码网络的非线性变化灰度差异图像配准[J]. 激光与光电子学进展, 2021, 58(16): 233-239.
	Huang Chao, Guo Hao, Gao Yan, et al. Nonlinear grayscale difference image registration based on stacked autoencoder network[J]. Laser & Optoelectronics Progress, 2021, 58(16): 233-239.
14	周丽丽, 朱佳琦, 王桥桥, 等. 基于雾线暗通道先验的水下图像复原方法[J].南京邮电大学学报: 自然科学版, 2020, 40(4): 64-69.
	Zhou Li-li, Zhu Jia-qi, Wang Qiao-qiao, et al. Underwater image restoration method based on haze-line and dark channel prior[J]. Journal of Nanjing University of Posts and Telecommunications (Natural Science), 2020, 40(4): 64-69.
15	胡秋生, 胡璋. 基于变分自编码器的非规则缺失图像修复仿真[J]. 计算机仿真, 2021, 38(12): 155-159.
	Hu Qiu-sheng, Hu Zhang. Simulation of irregular missing image restoration based on variational auto encoder[J]. Computer Simulation, 2021, 38(12): 155-159.
16	涂伟沪, 蔡玲霞, 李学军. 基于改进蝗虫算法优化Canny算子的鸡蛋裂纹图像检测[J]. 食品与机械, 2022, 38(2): 167-172, 202.
	Tu Wei-hu, Cai Ling-xia, Li Xue-jun. Egg crack image detection method based on improved grasshopper optimization algorithm and canny operator[J]. Food and Machinery, 2022, 38(2): 167-172, 202.
17	王煜玺, 袁晗, 朱宗晓, 等. 融合顶帽和Canny算子的电子束斑图像特征分析[J]. 中南民族大学学报: 自然科学版, 2022, 41(2): 222-227.
	Wang Yu-xi, Yuan Han, Zhu Zong-xiao, et al. Feature analysis of electron beam spot image by fusion of Top-hat and Canny operator[J]. Journal of South-Central University for Nationalities (Natural Science Edition), 2022, 41(2): 222-227.
18	吴孟礼, 陈跃斌, 吴海锋, 等. 抵御SSDF攻击的维纳滤波器检测算法研究[J]. 计算机工程, 2020, 46(11): 187-193.
	Wu Meng-li, Chen Yue-bin, Wu Hai-feng, et al. Research on wiener filter detection algorithm for resisting SSDF attacks[J]. Computer Engineering, 2020, 46(11): 187-193.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

算法	评价指标	图像a	图像b	图像c	图像d
本文	峰值信噪比/dB	70.6	72.4	71.8	75.3
本文	平均均方误差	0.25	0.24	0.22	0.23
文献［3］	峰值信噪比/dB	61.0	61.1	60.5	60.7
文献［3］	平均均方误差	0.51	0.50	0.53	0.54
文献［4］	峰值信噪比/dB	58.6	57.9	58.2	57.6
文献［4］	平均均方误差	0.46	0.51	0.49	0.53