A Method of Single-Image Light Source Interference Removal Based on the Multitask Network

被引：0

作者：

Zhang W. ^{[1
]}

Cheng G. ^{[1
]}

机构：

[1] School of Microelectronics, Tianjin University, Tianjin

来源：

Tianjin Daxue Xuebao (Ziran Kexue yu Gongcheng Jishu Ban)/Journal of Tianjin University Science and Technology | 2024年 / 57卷 / 05期

关键词：

deep learning; flare removal; image processing;

D O I：

10.11784/tdxbz202302023

中图分类号：

学科分类号：

摘要：

In the presence of light sources at a scene，captured images may be easily interfered with by the existence of various kinds of flares，glares，artifacts，etc. This leads to a lowered quality of key information，which adversely affects vision-based tasks such as object detection，depth estimation，and semantic segmentation based on those images. However，removing the light source interference of various forms and distributions with current methods is difficult. Therefore，we propose a light source interference removal network and its training method to solve issues regarding poor processing results，lack of paired training data，and weak generalization of existing algorithms. This network combines a multitasking structure to fully utilize the feature information of different tasks and thereby improves its interference removal performance. In addition，a postprocessing method for light source blend is presented to reduce the interference introduced in the postprocessing step. The proposed method realizes an average peak signal-to-noise ratio of 25.81 dB and 23.25 dB and a structural similarity of 0.872 6 and 0.922 3 when using public and self-built datasets，respectively. In terms of subjective qualitative comparison，the proposed method has better interference removal performance compared to existing methods. © 2024 Tianjin University. All rights reserved.

引用

页码：501 / 510

页数：9

共 28 条

[11] Qiao X，, Hancke G P，, Lau R., Light source guided single-image flare removal from unpaired data[C], 2021 IEEE/CVF International Conference on Computer Vision, pp. 4157-4165, (2021)
[12] Feng R C, Li C Y, Chen H J, Et al., Removing diffraction image artifacts in under-display camera via dynamic skip connection network[C], 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 662-671, (2021)
[13] Dai Y K, Li C Y, Et al., Flare7K：A phenomenological nighttime flare removal dataset[EB/OL]
[14] Zhou Y Y，, Liang D, Et al., Improving lens flare removal with general purpose pipeline and multiple light sources recovery[EB/OL]
[15] Cordts M，, Omran M, Et al., The cityscapes dataset for semantic urban scene understanding[C], 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp. 3213-3223, (2016)
[16] Zhang X E, Chen Q F., Single image reflection separation with perceptual losses[C], 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 4786-4794, (2018)
[17] Torralba A, Sinha P., Recognizing indoor scenes [C], 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 413-420, (2009)
[18] Armanfard N., Attentive task interaction network for multi-task learning[EB/OL]
[19] Shen Xiaohai, Li Zehao, Li Min, Et al., Aluminum surface-defect detection based on multi-task deep learning[J], Laser & Optoelectronics Progress, 57, 10, pp. 283-292, (2020)
[20] Yang Aiping, Cheng Simeng, Wang Jinbin, Et al., Saliency detection network based on edge detection and skeleton extraction[J], Journal of Tianjin University (Science and Technology), 56, 8, pp. 823-830, (2023)

← 1 2 3 →