Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing Aircraft

This paper presents FWStab, a specialized video stabilization dataset tailored for flapping-wing platforms. The dataset encompasses five typical flight scenarios, featuring 48 video clips with intense dynamic jitter. The corresponding Inertial Measurement Unit (IMU) sensor data are synchronously col...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhikai Wang, Sen Wang, Yiwen Hu, Yangfan Zhou, Na Li, Xiaofeng Zhang
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Biomimetics
Subjects:
Online Access:https://www.mdpi.com/2313-7673/10/7/448
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849303528403632128
author Zhikai Wang
Sen Wang
Yiwen Hu
Yangfan Zhou
Na Li
Xiaofeng Zhang
author_facet Zhikai Wang
Sen Wang
Yiwen Hu
Yangfan Zhou
Na Li
Xiaofeng Zhang
author_sort Zhikai Wang
collection DOAJ
description This paper presents FWStab, a specialized video stabilization dataset tailored for flapping-wing platforms. The dataset encompasses five typical flight scenarios, featuring 48 video clips with intense dynamic jitter. The corresponding Inertial Measurement Unit (IMU) sensor data are synchronously collected, which jointly provide reliable support for multimodal modeling. Based on this, to address the issue of poor image acquisition quality due to severe vibrations in aerial vehicles, this paper proposes a multi-modal signal fusion video stabilization framework. This framework effectively integrates image features and inertial sensor features to predict smooth and stable camera poses. During the video stabilization process, the true camera motion originally estimated based on sensors is warped to the smooth trajectory predicted by the network, thereby optimizing the inter-frame stability. This approach maintains the global rigidity of scene motion, avoids visual artifacts caused by traditional dense optical flow-based spatiotemporal warping, and rectifies rolling shutter-induced distortions. Furthermore, the network is trained in an unsupervised manner by leveraging a joint loss function that integrates camera pose smoothness and optical flow residuals. When coupled with a multi-stage training strategy, this framework demonstrates remarkable stabilization adaptability across a wide range of scenarios. The entire framework employs Long Short-Term Memory (LSTM) to model the temporal characteristics of camera trajectories, enabling high-precision prediction of smooth trajectories.
format Article
id doaj-art-b02eee7875cc4394bc2d14ecbc2993d5
institution Kabale University
issn 2313-7673
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Biomimetics
spelling doaj-art-b02eee7875cc4394bc2d14ecbc2993d52025-08-20T03:58:26ZengMDPI AGBiomimetics2313-76732025-07-0110744810.3390/biomimetics10070448Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing AircraftZhikai Wang0Sen Wang1Yiwen Hu2Yangfan Zhou3Na Li4Xiaofeng Zhang5College of Information Engineering, Henan University of Science and Technology, Luoyang 471023, ChinaCollege of Information Engineering, Henan University of Science and Technology, Luoyang 471023, ChinaNational Key Laboratory of Digital and Agile Aircraft Design, Chengdu 610091, ChinaSuzhou Institute of Nano-Tech and Nano-Bionics (SINANO), Chinese Academy of Sciences, Suzhou 215123, ChinaCollege of Information Engineering, Henan University of Science and Technology, Luoyang 471023, ChinaSchool of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin 150001, ChinaThis paper presents FWStab, a specialized video stabilization dataset tailored for flapping-wing platforms. The dataset encompasses five typical flight scenarios, featuring 48 video clips with intense dynamic jitter. The corresponding Inertial Measurement Unit (IMU) sensor data are synchronously collected, which jointly provide reliable support for multimodal modeling. Based on this, to address the issue of poor image acquisition quality due to severe vibrations in aerial vehicles, this paper proposes a multi-modal signal fusion video stabilization framework. This framework effectively integrates image features and inertial sensor features to predict smooth and stable camera poses. During the video stabilization process, the true camera motion originally estimated based on sensors is warped to the smooth trajectory predicted by the network, thereby optimizing the inter-frame stability. This approach maintains the global rigidity of scene motion, avoids visual artifacts caused by traditional dense optical flow-based spatiotemporal warping, and rectifies rolling shutter-induced distortions. Furthermore, the network is trained in an unsupervised manner by leveraging a joint loss function that integrates camera pose smoothness and optical flow residuals. When coupled with a multi-stage training strategy, this framework demonstrates remarkable stabilization adaptability across a wide range of scenarios. The entire framework employs Long Short-Term Memory (LSTM) to model the temporal characteristics of camera trajectories, enabling high-precision prediction of smooth trajectories.https://www.mdpi.com/2313-7673/10/7/448biomimetic aircraftcross-attention mechanismoptical flowSEA-RAFTimage enhancement
spellingShingle Zhikai Wang
Sen Wang
Yiwen Hu
Yangfan Zhou
Na Li
Xiaofeng Zhang
Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing Aircraft
Biomimetics
biomimetic aircraft
cross-attention mechanism
optical flow
SEA-RAFT
image enhancement
title Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing Aircraft
title_full Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing Aircraft
title_fullStr Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing Aircraft
title_full_unstemmed Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing Aircraft
title_short Multimodal Fusion Image Stabilization Algorithm for Bio-Inspired Flapping-Wing Aircraft
title_sort multimodal fusion image stabilization algorithm for bio inspired flapping wing aircraft
topic biomimetic aircraft
cross-attention mechanism
optical flow
SEA-RAFT
image enhancement
url https://www.mdpi.com/2313-7673/10/7/448
work_keys_str_mv AT zhikaiwang multimodalfusionimagestabilizationalgorithmforbioinspiredflappingwingaircraft
AT senwang multimodalfusionimagestabilizationalgorithmforbioinspiredflappingwingaircraft
AT yiwenhu multimodalfusionimagestabilizationalgorithmforbioinspiredflappingwingaircraft
AT yangfanzhou multimodalfusionimagestabilizationalgorithmforbioinspiredflappingwingaircraft
AT nali multimodalfusionimagestabilizationalgorithmforbioinspiredflappingwingaircraft
AT xiaofengzhang multimodalfusionimagestabilizationalgorithmforbioinspiredflappingwingaircraft