A Hybrid Moving Object Detection Method for Aerial Images

Huang, Chung-Hsien; Wu, You; Kao, Jau-Hong; Shih, Ming-Yu; Chou, Cheng-Chuan

doi:10.1007/978-3-642-15702-8_33

Cited by 25 publications

(14 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Global registration methods used for aerial image registration include mutual information [14] which corrects only the translation, and region phase correlation [17] which estimates 8-DOF homography. Feature based methods include a wide gamut of approaches as those using Harris corners [9] [2], SURF features [1], SIFT features [11] [16] and Shi & Tomasi corners [5]. Approaches found to combine both techniques use SIFT features with Mutual Information [11], Harris corners with Efficient Second order Minimization [15] and Harris corners with gradient based alignment [2].…”

Section: Related Workmentioning

confidence: 99%

“…Once the camera motion compensation step is performed, the moving objects can be detected by: frame differencing [5][1] [3], accumulative frame differencing [4] [9] [2], median background subtraction, statistical mode background subtraction [11], or normal optical flow [14]. Since frame differencing techniques do not segment the whole object, image segmentation techniques are used to improve the results [1] [9].…”

Section: Related Workmentioning

confidence: 99%

“…Since frame differencing techniques do not segment the whole object, image segmentation techniques are used to improve the results [1] [9]. Probabilistic background modeling, such as Gaussian Mixture Model (GMM) are avoided according to [4] for the following reasons: the frame rate is not high enough to learn the rapidly changing background, there is an accumulated alignment error because of consecutive homography computations, and there are errors produced due to parallax.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Telemetry assisted frame registration and background subtraction in low-altitude UAV videos

Tzanidou

Climent-Pérez

Hummel

et al. 2015

2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

View full text Add to dashboard Cite

This work presents an approach to detect moving objects from Unmanned Aerial Vehicles (UAV). A common framework for most of the existing techniques is using image registration to warp consecutive frames as an ego-motion compensation step and applying frame differencing to detect the moving objects. Assuming a planar scene, we propose the exploitation of telemetry information available from Global Positioning and Inertial Navigation Systems (GPS/INS) to estimate a similarity transformation matrix that would map the image points from one frame to another. In this work, we show that the telemetry-based image registration combined with global registration methods produces more accurate results than the traditional image registration techniques in case of a scene with poor or no texture. To segment the moving objects, we employ the probabilistic background modelling method with mixture of Gaussian distributions.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Telemetry assisted frame registration and background subtraction in low-altitude UAV videos

Tzanidou

Climent-Pérez

Hummel

et al. 2015

2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

View full text Add to dashboard Cite

show abstract

“…As shown by Miller et al [19], one possible approach is to use background subtraction methods with a combination of intensity threshold (for IR imagery), motion compensation and pattern classification. Chung et al [20] applied accumulative frame differencing to detect the pixels with motion and combined these pixels with homogeneous regions in the frame obtained by image segmentation. Other methods use optical flow as the main analysis technique.…”

Section: Introductionmentioning

confidence: 99%

“…Herein, we have developed a new method that combines egomotion determination based on static point features with optical flow comparison to determine pixels that belong to dynamic objects. Chung et al [20] proposed a frame differencing procedure that would not work in our case due to the high frequency vibrations in the movement of currently available commercial UAVs. Meanwhile, our method only tracks single static features to determine the movement of the camera.…”

Section: Introductionmentioning

confidence: 99%

A Real-Time Method to Detect and Track Moving Objects (DATMO) from Unmanned Aerial Vehicles (UAVs) Using a Single Camera

et al. 2012

View full text Add to dashboard Cite

We develop a real-time method to detect and track moving objects (DATMO) from unmanned aerial vehicles (UAVs) using a single camera. To address the challenging characteristics of these vehicles, such as continuous unrestricted pose variation and low-frequency vibrations, new approaches must be developed. The main concept proposed in this work is to create an artificial optical flow field by estimating the camera motion between two subsequent video frames. The core of the methodology consists of comparing this artificial flow with the real optical flow directly calculated from the video feed. The motion of the UAV between frames is estimated with available parallel tracking and mapping techniques that identify good static features in the images and follow them between frames. By comparing the two optical flows, a list of dynamic pixels is obtained and then grouped into dynamic objects. Tracking these dynamic objects through time and space provides a filtering procedure to eliminate spurious events and misdetections. The algorithms have been tested with a quadrotor platform using a commercial camera

show abstract

Real‐time visual tracking design for an unmanned aerial vehicle in cluttered environments

Liang,

Yi,

et al. 2024

Optim Control Appl Methods

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) are widely used in various fields, and autonomous tracking of UGV by UAV can significantly improve the collaborative capability and operational range of unmanned systems. In order to realize autonomous UAV tracking in complex and cluttered scenarios, this paper presents a vision‐based tracking system for UAV tracking UGV, integrating a visual target tracking model based on deep learning with a local path planning methodology. The system includes a deep learning‐based tracker, a state machine, and an obstacle avoidance planning algorithm. In addition, to improve the robustness of UAV onboard tracking, we introduce a lightweight tracking algorithm based on MobileNetV2 network, which ensures the tracking performance while improving the operation speed. Through real‐world experiments, it is demonstrated that our system can realize autonomous target tracking in cluttered environments without prior information. The proposed tracker exhibits a success rate of 61.7% on the UAV123 dataset and achieves a tracking speed of 45 frames per second (fps) on the NVIDIA Jetson TX2, demonstrating significant advancements in real‐time, efficient unmanned tracking technologies.

show abstract

A Hybrid Moving Object Detection Method for Aerial Images

Cited by 25 publications

References 13 publications

Telemetry assisted frame registration and background subtraction in low-altitude UAV videos

Telemetry assisted frame registration and background subtraction in low-altitude UAV videos

A Real-Time Method to Detect and Track Moving Objects (DATMO) from Unmanned Aerial Vehicles (UAVs) Using a Single Camera

Real‐time visual tracking design for an unmanned aerial vehicle in cluttered environments

Contact Info

Product

Resources

About