Adaptive smoothing in real-time image stabilization

Wu, Shaomin; Zhang, David C.; Zhang, Yuzheng; Basso, James; Melle, Michael T.

doi:10.1117/12.918737

Cited by 5 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In motion smoothing, high‐frequency jitters are damped from the accumulative global motion. The commonly used motion smoothing methods include spline smoothing [21, 22], feature trajectory smoothing [23, 24], motion vector integration (MVI) [25], Gaussian filter [26], Kalman filter (KF) [13], and adaptive filters [27]. Motion smoothing is not a trivial process as the extent of intentional motion varies in video sequences captured from diverse platforms.…”

Section: Introductionmentioning

confidence: 99%

Robust and computationally efficient online image stabilisation framework based on adaptive dual motion vector integration

Yousaf

Hanif

Khan

et al. 2019

IET Computer Vision

View full text Add to dashboard Cite

Image stabilisation aims to compensate and smoothen the effects of undesired trembling motion of cameras mounted on non-static platforms. It becomes quite a challenging task in the case of moving platforms, such as ground vehicles, unmanned aerial vehicles, and handheld devices. Many satisfactory solutions to the image stabilisation problem are proposed in the recent literature, but most of these methods are not adaptable for handling a wide range of intentional motions with minimum lag, especially in real-time scenarios. In this study, the authors propose an online two-dimensional image stabilisation technique based on dual motion vector integration, which is a novel adaptive motion smoothing technique that employs an average length of motion vectors to estimate the intentional motion. The overall computational cost of the proposed system is significantly reduced by employing frame-shaking judgment that only allows processing of jittering frames. Promising experimental results have been obtained on challenging videos obtained from hand-held and vehicle-mounted cameras which demonstrate the robustness and effectiveness of the proposed technique against feature point mismatching and presence of moving objects within the scene at a frame rate of 30 frames per second.

show abstract

Section: Introductionmentioning

confidence: 99%

Robust and computationally efficient online image stabilisation framework based on adaptive dual motion vector integration

Yousaf

Hanif

Khan

et al. 2019

IET Computer Vision

View full text Add to dashboard Cite

show abstract

“…Several motion smoothing methods are available for motion intention estimation such as particle filter [10], Kalman filter [11], Gaussian filter [25,26], adaptive filter [26,27], spline smoothing [28,29], or point feature trajectory smoothing [30,31]. In our approach, the control signal sent to the MAV is dealt as a known information; hence, a new and different methodology to those in the literature is considered.…”

Section: Motion Intention Estimationmentioning

confidence: 99%

Real-time video stabilization without phantom movements for micro aerial vehicles

Aguilar

Ángulo

2014

J Image Video Proc

View full text Add to dashboard Cite

In recent times, micro aerial vehicles (MAVs) are becoming popular for several applications as rescue, surveillance, mapping, etc. Undesired motion between consecutive frames is a problem in a video recorded by MAVs. There are different approaches, applied in video post-processing, to solve this issue. However, there are only few algorithms able to be applied in real time. An additional and critical problem is the presence of false movements in the stabilized video. In this paper, we present a new approach of video stabilization which can be used in real time without generating false movements. Our proposal uses a combination of a low-pass filter and control action information to estimate the motion intention.

show abstract

“…Múltiples métodos de suavizado de movimiento se encuentran disponibles para la estimación de la intención de movimiento como particle filter [49], Kalman filter [82], Gaussian filter [40], [30]. adaptive filter [87], [88], spline smoothing [89], [90], o point feature trajectory smoothing [91], [92].…”

Section: Estimación De La Intención De Movimientounclassified

Estabilización de vídeo en tiempo real : aplicaciones en teleoperación de micro vehículos aéreos de ala rotativa

Aguilar Castillo

View full text Add to dashboard Cite

Micro Aerial Vehicles (MAVs), a subset of Unmanned Aerial Vehicles (UAVs), also known as drones, are becoming popular for several applications and gaining interest due to advantages as manufacturing and maintenance cost, size and weight, energy consumption, and flight maneuverability. Required skills for drone teleoperators being lower than for aircraft pilots, however their training process can last several weeks or months depending on the target at hands. In particular, this process is harder when teleoperators cannot observe directly the vehicle, depending only on onboard sensors and cameras. The presence of oscillations in the captured video is a major problem with cameras on UAVs. It is even more complex for MAVs because the external disturbances increase the instability. There exists mechanical video stabilizers that reduce camera oscillations, however this mechanical device adds weight and increases the manufacturing cost, energy consumption, size, weight, and the system becomes less safe for people. In this thesis, we propose to develop video stabilization software algorithms, without additional mechanical elements in the system, to be applied in real-time during the UAV navigation. In the literature, there are a few video stabilization algorithms able to be applied in real-time, but most of them generate false motion (phantom movements) in the stabilized image. Our algorithm represents a good tradeoff between stable video recording and simultaneously keeping UAV real motion. Several experiments with MAVs have been performed and the employed measurements demonstrate the good performance of the introduced algorithm. Los micro vehículos aéreos (MAVs), un subconjunto de vehículos aéreos no tripulados (UAVs), también llamados drones, han ganado popularidad en múltiples aplicaciones y un creciente interés debido a sus ventajas como costo de fabricación y mantenimiento, volumen, peso del vehículo, gasto energético, y maniobrabilidad de vuelo. La destreza requerida para un teleoperador de drones es inferior a la de un piloto de aeronaves de mayor dimensión, no obstante, su proceso de entrenamiento puede durar varias semanas o incluso meses dependiendo del objetivo que se persiga. Este proceso se dificulta cuando el teleoperador no puede observar de forma directa al vehículo y depende únicamente de los sensores y cámaras a bordo del sistema. Uno de los principales problemas con cámaras a bordo de drones es la oscilación presente en los vídeos capturados. Este inconveniente es más complejo para los MAVs porque las perturbaciones externas provocan mayor inestabilidad. Existen dispositivos mecánicos de estabilización de vídeo que reducen las oscilaciones en la cámara. Sin embargo, estos mecanismos implican una carga adicional al sistema y aumentan el costo de producción, gasto energético y el riesgo para las personas que se encuentren cerca en caso de accidente. En la presente tesis se propone el desarrollo de algoritmos de estabilización de vídeo por software sin elementos mecánicos adicionales en el sistema, a ser utilizados en tiempo real durante la navegación de los UAVs. En la literatura existen pocos algoritmos de estabilización de video aplicables en tiempo real, los cuales generan falsos movimientos (movimientos fantasma) en la imagen estabilizada. El algoritmo desarrollado es capaz de obtener una imagen estable y simultáneamente mantener los movimientos reales. Se han llevado a cabo múltiples experimentos con MAVs y las métricas de evaluación utilizadas evidencian el buen desempeño del algoritmo introducido.

show abstract

Adaptive smoothing in real-time image stabilization

Cited by 5 publications

References 0 publications

Robust and computationally efficient online image stabilisation framework based on adaptive dual motion vector integration

Robust and computationally efficient online image stabilisation framework based on adaptive dual motion vector integration

Real-time video stabilization without phantom movements for micro aerial vehicles

Estabilización de vídeo en tiempo real : aplicaciones en teleoperación de micro vehículos aéreos de ala rotativa

Contact Info

Product

Resources

About