基于小波和动态互补滤波的图像与事件融合方法
TP391; 本文探讨了事件相机和传统帧相机数据的融合,引入了一种旨在增强复杂光照条件下成像质量的新型融合方法.事件相机作为创新性视觉传感器的代表,以其高时域分辨率和极低功耗受到广泛关注.然而,事件相机产生的数据往往存在噪声和特征丢失等问题.相反,传统帧相机具有较高的空间分辨率,但在捕捉涉及快速运动或包含广泛动态范围场景时表现不佳.基于此,本文引入了一种创新方法,将离散小波变换与动态增益互补滤波相结合,有效地融合图像和事件数据.该方法首先通过计算图像的信息熵评估曝光水平.然后,采用离散小波变换来从事件数据和帧图像数据中分离高频和低频细节.随后,该方法应用动态增益互补滤波器来实现图像和事件数据的...
Saved in:
| Published in | 工程科学学报 Vol. 46; no. 11; pp. 2076 - 2084 |
|---|---|
| Main Authors | , , |
| Format | Journal Article |
| Language | Chinese |
| Published |
北京控制工程研究所,北京 100190
01.11.2024
|
| Subjects | |
| Online Access | Get full text |
| ISSN | 2095-9389 |
| DOI | 10.13374/j.issn2095-9389.2024.01.23.004 |
Cover
| Abstract | TP391; 本文探讨了事件相机和传统帧相机数据的融合,引入了一种旨在增强复杂光照条件下成像质量的新型融合方法.事件相机作为创新性视觉传感器的代表,以其高时域分辨率和极低功耗受到广泛关注.然而,事件相机产生的数据往往存在噪声和特征丢失等问题.相反,传统帧相机具有较高的空间分辨率,但在捕捉涉及快速运动或包含广泛动态范围场景时表现不佳.基于此,本文引入了一种创新方法,将离散小波变换与动态增益互补滤波相结合,有效地融合图像和事件数据.该方法首先通过计算图像的信息熵评估曝光水平.然后,采用离散小波变换来从事件数据和帧图像数据中分离高频和低频细节.随后,该方法应用动态增益互补滤波器来实现图像和事件数据的融合.该方法的核心在于其能够自适应地平衡每种数据源的贡献,从而确保在不同条件下实现最佳的图像重建质量.通过利用事件相机的高频时域信息和帧相机的高分辨率空间信息,该方法旨在克服每种传感器固有的局限性.这种融合不仅可以减轻事件相机数据中遇到的噪声和特征丢失,还可以解决帧相机在捕捉高速运动和具有明显亮度变化的场景时的缺点.该融合方法的有效性已经在HDR Hybrid Event-Frame数据集上进行了测试,该数据集具有高动态范围和复杂光照环境的真实场景.实验结果展示了该方法在提高图像质量方面实现的改进.与传统图像重建方法相比,本文方法在几个关键指标上表现出色:均方误差为0.0199,结构相似性指数为0.90,Q-score为6.07.这些结果不仅验证了所提出的融合方法在改善在挑战性条件下成像质量方面的有效性,还凸显了整合不同类型视觉数据以实现更优重建结果的潜力. |
|---|---|
| AbstractList | TP391; 本文探讨了事件相机和传统帧相机数据的融合,引入了一种旨在增强复杂光照条件下成像质量的新型融合方法.事件相机作为创新性视觉传感器的代表,以其高时域分辨率和极低功耗受到广泛关注.然而,事件相机产生的数据往往存在噪声和特征丢失等问题.相反,传统帧相机具有较高的空间分辨率,但在捕捉涉及快速运动或包含广泛动态范围场景时表现不佳.基于此,本文引入了一种创新方法,将离散小波变换与动态增益互补滤波相结合,有效地融合图像和事件数据.该方法首先通过计算图像的信息熵评估曝光水平.然后,采用离散小波变换来从事件数据和帧图像数据中分离高频和低频细节.随后,该方法应用动态增益互补滤波器来实现图像和事件数据的融合.该方法的核心在于其能够自适应地平衡每种数据源的贡献,从而确保在不同条件下实现最佳的图像重建质量.通过利用事件相机的高频时域信息和帧相机的高分辨率空间信息,该方法旨在克服每种传感器固有的局限性.这种融合不仅可以减轻事件相机数据中遇到的噪声和特征丢失,还可以解决帧相机在捕捉高速运动和具有明显亮度变化的场景时的缺点.该融合方法的有效性已经在HDR Hybrid Event-Frame数据集上进行了测试,该数据集具有高动态范围和复杂光照环境的真实场景.实验结果展示了该方法在提高图像质量方面实现的改进.与传统图像重建方法相比,本文方法在几个关键指标上表现出色:均方误差为0.0199,结构相似性指数为0.90,Q-score为6.07.这些结果不仅验证了所提出的融合方法在改善在挑战性条件下成像质量方面的有效性,还凸显了整合不同类型视觉数据以实现更优重建结果的潜力. |
| Abstract_FL | This study investigates the fusion of data from event cameras and traditional frame cameras,introducing a novel fusion approach designed to enhance image quality under complex lighting conditions.Event cameras are an innovative class of vision sensors that are known for their high temporal resolution and minimal power consumption;however,their output is often plagued by noise and feature loss.Conversely,traditional frame cameras boast commendable spatial resolution;however,they struggle to capture fast-moving scenes or scenes with a vast dynamic range.To address these challenges,the study proposes an innovative method that combines discrete wavelet transform with dynamic gain complementary filtering to fuse image and event data.The process begins by evaluating the exposure level of incoming image frames using the image entropy of a metric.Following this assessment,the discrete wavelet transform segregates the high-and low-frequency components from the event stream and frame image data.A dynamic gain complementary filter is applied to seamlessly integrate image and event data.The proposed method capitalizes on its ability to balance the contribution of each data source adaptively,thereby ensuring optimal reconstruction quality under varying conditions.By leveraging the high-frequency temporal information from event cameras and the high-resolution spatial information from frame cameras,the proposed method attempts to overcome the limitations inherent in each type of sensor.This fusion not only mitigates the noise and feature loss in event camera data but also improves the capture of high-speed movements and scenes with significant brightness variations.The efficacy of this fusion approach was rigorously tested on the HDR Hybrid Event-Frame Dataset,which includes high dynamic range and complex lighting environments in real-world scenarios.The experimental results underscored a notable improvement in image quality,outperforming traditional image reconstruction methods.Our proposed approach has demonstrated superior performance,as evidenced by its scores on several key metrics:a mean squared error of 0.0199,a structural similarity index measure of 0.90,and a Q-score of 6.07.These results not only validate the effectiveness of the proposed fusion method in enhancing imaging quality under challenging conditions but also highlight the potential of integrating disparate types of visual data to achieve superior reconstruction outcomes. |
| Author | 王瑞琳 贺盈波 王立 |
| AuthorAffiliation | 北京控制工程研究所,北京 100190 |
| AuthorAffiliation_xml | – name: 北京控制工程研究所,北京 100190 |
| Author_FL | WANG Ruilin HE Yingbo WANG Li |
| Author_FL_xml | – sequence: 1 fullname: WANG Ruilin – sequence: 2 fullname: WANG Li – sequence: 3 fullname: HE Yingbo |
| Author_xml | – sequence: 1 fullname: 王瑞琳 – sequence: 2 fullname: 王立 – sequence: 3 fullname: 贺盈波 |
| BookMark | eNo9j81Kw1AUhO-igrX2OcRF4jn3nDTJSqT4BwU3ui653kQaJQWD2KWiC_GnlYIuFBRFwaVQEE3Bp8mNfQwriquBGfhmZkqUknYSCjGDYCORy3Ox3UrTRILvWD55vi1Bsg1oS7IBuCTK_9GkqKZpS4GD5KIvoSzmzX2WZ13z2isGj6Z_bk5fioPDPOuPHp6L4dPY_Lo5Nref5qiXv3fz7Cwfvo3uLszlSXH9UQyupsVEFOykYfVPK2JjaXG9vmI11pZX6wsNK0VgtjyqMYNLUU1JcEFqDp1Qe-MJge9RCEAEEUWsdIBSIeKmL5lRRw67WmFIFTH7y90PkihItppxe283GTc2Vbwd605H_bxGBGT6Bg6xZiw |
| ClassificationCodes | TP391 |
| ContentType | Journal Article |
| Copyright | Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| Copyright_xml | – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| DBID | 2B. 4A8 92I 93N PSX TCJ |
| DOI | 10.13374/j.issn2095-9389.2024.01.23.004 |
| DatabaseName | Wanfang Data Journals - Hong Kong WANFANG Data Centre Wanfang Data Journals 万方数据期刊 - 香港版 China Online Journals (COJ) China Online Journals (COJ) |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| DocumentTitle_FL | Image and event fusion method based on wavelet and dynamic complementary filtering |
| EndPage | 2084 |
| ExternalDocumentID | bjkjdxxb202411014 |
| GrantInformation_xml | – fundername: 国防科技卓越青年科学基金资助项目 funderid: (2021-JCJQ-ZQ-045) |
| GroupedDBID | -0C -SC -S~ 2B. 2RA 4A8 5VR 92I 92M 93N 9D9 9DC AAITT AFUIB ALMA_UNASSIGNED_HOLDINGS CAJEC CQIGP FA0 GROUPED_DOAJ JUIAU PB1 PB6 PSX Q-- Q-2 R-C RT3 T8S TCJ U1F U5C |
| ID | FETCH-LOGICAL-s1044-83644073f6b20702d4e5ed8920a983e00330f3f4bda12b111c92441df547db1e3 |
| ISSN | 2095-9389 |
| IngestDate | Thu May 29 04:07:32 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Issue | 11 |
| Keywords | event camera 小波变换 图像重建 complementary filtering image and event fusion 互补滤波 事件相机 wavelet transform image reconstruction 图像与事件融合 |
| Language | Chinese |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-s1044-83644073f6b20702d4e5ed8920a983e00330f3f4bda12b111c92441df547db1e3 |
| PageCount | 9 |
| ParticipantIDs | wanfang_journals_bjkjdxxb202411014 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-11-01 |
| PublicationDateYYYYMMDD | 2024-11-01 |
| PublicationDate_xml | – month: 11 year: 2024 text: 2024-11-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | 工程科学学报 |
| PublicationTitle_FL | Chinese Journal of Engineering |
| PublicationYear | 2024 |
| Publisher | 北京控制工程研究所,北京 100190 |
| Publisher_xml | – name: 北京控制工程研究所,北京 100190 |
| SSID | ssib051371920 ssib023167159 ssj0003313525 ssib022319478 ssib041261352 ssib036435564 |
| Score | 2.3782024 |
| Snippet | TP391;... |
| SourceID | wanfang |
| SourceType | Aggregation Database |
| StartPage | 2076 |
| Title | 基于小波和动态互补滤波的图像与事件融合方法 |
| URI | https://d.wanfangdata.com.cn/periodical/bjkjdxxb202411014 |
| Volume | 46 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnR1Na9RAdGgriBdRVPymiIMH2Zr5SGbmJMk2S_HgqYXeSmaTKBVWsFsoPSl6ED9aKehBQVEUPAoF0S34azZrf4bvTbK7gS5SvYS3M-8zL5n3Mjszj5CrvlYJRA3eYCnLGhCvZSPJ4adnrPJ5Yk2QuVW-t4OFJXlr2V-emra1VUvrXTvX3py4r-R_vApt4FfcJfsPnh0xhQaAwb9wBQ_D9VA-prFPTYtGIY0lXnWMLZFHdYvGAY0EDbnD4VQ3EdAhDTV2acBhFRX0xpqGjIa-o4poKGvkihrgLB2fiEZOhBZOBJBrJ7SUHjkAcAJkaGKq5x0VyHJCDfA0Feey5uUwLXZqK6eAQg2RFQAgmmFXOE_DoAYEzhB_-LAgLmhREgEFSEbAAzkTUcIIgHGPppHTH2miSlVne31GhMtqa-DwGXa3oUmNqqwPm06vGLXGLu1uw0SzPGqkA4zDASqDDuHNgzyv4-FVZaXTcrjmHla9FGVBpGFsqaZXq3eI1SOFp4Ja1sG9slLegYgmhJIupKGQkYw5tBvPm-V4QK8cB_PREku7em813diwiMiwIPM0OcKVKpcyVNMOMOZCdshM7Ug3jicj1FJcARmrXztPSDL45Bb-KCX2mVBsWPAMsx8hsBtXDo-UPUquDS258Xc73M65Tp507tSSvMUT5Hj1dTYblq_aSTK1efcUuVl86PV7W8W37cHup2LnRfHs6-Dho35vZ__jl8HeZ2j8_fZJ8e5X8Xi7_2Or33ve3_u-__5l8erp4M3Pwe7r02SpFS82FxpV3ZHGGvOkbGgwWkLoywO4e8rjqcz8LNVgZWK0yNBGLxe5tGnCuIVkoW1guGNp7kuVWpaJM2Smc7-TnSWz2iov89ttj8EImAXcJrhzxsLwmOAf1vwcuVIZvFKNK2srBzx3_jBIF8ix8btwkcx0H6xnlyBf7trLzuF_AKhDjnk |
| linkProvider | Directory of Open Access Journals |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8E%E5%B0%8F%E6%B3%A2%E5%92%8C%E5%8A%A8%E6%80%81%E4%BA%92%E8%A1%A5%E6%BB%A4%E6%B3%A2%E7%9A%84%E5%9B%BE%E5%83%8F%E4%B8%8E%E4%BA%8B%E4%BB%B6%E8%9E%8D%E5%90%88%E6%96%B9%E6%B3%95&rft.jtitle=%E5%B7%A5%E7%A8%8B%E7%A7%91%E5%AD%A6%E5%AD%A6%E6%8A%A5&rft.au=%E7%8E%8B%E7%91%9E%E7%90%B3&rft.au=%E7%8E%8B%E7%AB%8B&rft.au=%E8%B4%BA%E7%9B%88%E6%B3%A2&rft.date=2024-11-01&rft.pub=%E5%8C%97%E4%BA%AC%E6%8E%A7%E5%88%B6%E5%B7%A5%E7%A8%8B%E7%A0%94%E7%A9%B6%E6%89%80%2C%E5%8C%97%E4%BA%AC+100190&rft.issn=2095-9389&rft.volume=46&rft.issue=11&rft.spage=2076&rft.epage=2084&rft_id=info:doi/10.13374%2Fj.issn2095-9389.2024.01.23.004&rft.externalDocID=bjkjdxxb202411014 |
| thumbnail_s | http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fbjkjdxxb%2Fbjkjdxxb.jpg |