Detecting and tracking dim small targets in infrared image sequences under complex backgrounds

Li, Ying; Shi, Liang; Bai, Bendu; Feng, David Dagan

doi:10.1007/s11042-012-1258-y

Cited by 31 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As for the initial position of this procedure, in the first frame a rectangular patch centered at is manually initialized as the target model, and the iterations begin at . For the rest of the frames, the target position of the previous frame is used as the initial position of the current frame [22]. This method requires the target regions in consecutive two frames having some overlap areas.…”

Section: Theory Of Mean Shift Trackingmentioning

confidence: 99%

Extended social force model‐based mean shift for pedestrian tracking under obstacle avoidance

Zhang

Wang³

et al. 2016

IET Computer Vision

View full text Add to dashboard Cite

It has been shown that the mean shift tracking algorithm can achieve excellent results in the task of pedestrian tracking. It empirically estimates the target position of the current frame by locating the maximum of a density function from the local neighborhood of the target position of the previous frame. However, this method only considers its past trajectory without taking into account the influence of the pedestrian's environment. In practice, pedestrians always keep a safe distance away from obstacles when programming their paths. To address the issue of obstacle avoidance, this paper proposes a novel extended social force model-based mean shift tracking algorithm, in which the pedestrian's environment is taken into full consideration. Firstly, in order to show how the environment impacts pedestrian movements from the viewpoint of force, an extended social force model is presented by considering the interaction between target and obstacle. Furthermore, according to characteristics of pedestrian tracking, directional weights and speed weights are introduced to adjust the strength of the force concerning the difference of individual perspectives and relative velocities. Finally, the initial target position is predicted by Newton's laws of motion and then the mean shift method is integrated to track the target position. Experiment results showed that this algorithm achieved an encouraging performance when an obstacle occurred. An object that moves fast or changes its moving directions quickly can also be robustly tracked in real time by using the proposed algorithm.

show abstract

Section: Theory Of Mean Shift Trackingmentioning

confidence: 99%

Extended social force model‐based mean shift for pedestrian tracking under obstacle avoidance

Zhang

Wang³

et al. 2016

IET Computer Vision

View full text Add to dashboard Cite

show abstract

“…To achieve a satisfying target detection performance, many approaches have been proposed for different scenarios, including two types: Track-before-detection (TBD) approaches [4,5] and detection-before-track (DBT) approaches [6][7][8]. TBD approaches have good detection performance for targets with continuous track motion, such as 3D matched filters [9] and its improved versions [10,11].…”

Section: Introductionmentioning

confidence: 99%

Infrared Small-Faint Target Detection Using Non-i.i.d. Mixture of Gaussians and Flux Density

Sun

Yang

et al. 2019

Remote Sensing

View full text Add to dashboard Cite

The robustness of infrared small-faint target detection methods to noisy situations has been a challenging and meaningful research spot. The targets are usually spatially small due to the far observation distance. Considering the underlying assumption of noise distribution in the existing methods is impractical; a state-of-the-art method has been developed to dig out valuable information in the temporal domain and separate small-faint targets from background noise. However, there are still two drawbacks: (1) The mixture of Gaussians (MoG) model assumes that noise of different frames satisfies independent and identical distribution (i.i.d.); (2) the assumption of Markov random field (MRF) would fail in more complex noise scenarios. In real scenarios, the noise is actually more complicated than the MoG model. To address this problem, a method using the non-i.i.d. mixture of Gaussians (NMoG) with modified flux density (MFD) is proposed in this paper. We firstly construct a novel data structure containing spatial and temporal information with an infrared image sequence. Then, we use an NMoG model to describe the noise, which can be separated with the background via the variational Bayes algorithm. Finally, we can select the component containing true targets through the obvious difference of target and noise in an MFD maple. Extensive experiments demonstrate that the proposed method performs better in complicated noisy scenarios than the competitive approaches.

show abstract

“…Infrared (IR) target tracking and detecting play a crucial role in military, video surveillance, and aeronautics applications [1,2]. In military applications, IR imaging guidance is the development trend of IR guidance technology, which has attracted considerable attention due to its outstanding advantages of strong anti-interference ability, long detection distance, all-weather observations, and high guidance precision [3]. However, in contrast to visual images, IR images generally have low spatial resolution, poor signal-to-noise (SNR) ratios, and lack of textural information (see Figure 1) [4].…”

Section: Introductionmentioning

confidence: 99%

“…Extensive work has been done in the area of aerial IR target tracking [3,[6][7][8]. Meanwhile, numerous superior trackers have been proposed in visual tracking, especially the deep tracker.…”

Section: Introductionmentioning

confidence: 99%

Aerial Infrared Target Tracking in Complex Background Based on Combined Tracking and Detecting

Xiao

Zhang

et al. 2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Aerial infrared target tracking is the basis of many weapon systems, especially the air-to-air missile. Till now, it is still challenging research to track the aircraft in the event of complex background. In this paper, we focus on developing an algorithm that could track the aircraft fast and accurately based on infrared image sequence. We proposed a framework composed of a tracker based on correlation filter and a detector based on deep learning, which we call combined tracking and detecting (CTAD). With such collaboration, the algorithm enjoys both the high efficiency provided by correlation filter and the strong discriminative power provided by deep learning. Finally, we performed experiments on three representative infrared image sequences and two sequences from VOT-TIR2016 dataset to quantitatively evaluate the performance of our algorithm. To evaluate our algorithm scientifically, we present the experiments performed on two sequences from AMCOM FLIR dataset of the proposed algorithm. The experimental results demonstrate that our algorithm could track the infrared target reliably, which shows comparable performance with the deep tracker, while running at a fast speed of about 18.1 fps.

show abstract

Detecting and tracking dim small targets in infrared image sequences under complex backgrounds

Cited by 31 publications

References 28 publications

Extended social force model‐based mean shift for pedestrian tracking under obstacle avoidance

Extended social force model‐based mean shift for pedestrian tracking under obstacle avoidance

Infrared Small-Faint Target Detection Using Non-i.i.d. Mixture of Gaussians and Flux Density

Aerial Infrared Target Tracking in Complex Background Based on Combined Tracking and Detecting

Contact Info

Product

Resources

About