DeepDA: LSTM-based Deep Data Association Network for Multi-Targets Tracking in Clutter

Liu, Huajun; Zhang, Hui; Mertz, Christoph

doi:10.23919/fusion43075.2019.9011217

Cited by 22 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The LSTM model can deal with sequence images and is effective for nonlinear changing targets, so data association based on LSTM has been applied to linear distribution. 42 The results of step-by-step prediction probabilities and measurements show that the the method performs more stable compare to traditional methods. Hierarchical association method is widely used, and many technologies have been put forward, which has advantages in accuracy, robustness and speed, but its outstanding disadvantage is that the computational complexity of the network will increase when splitting tasks or adding layers between multiple association levels.…”

Section: Other Methodsmentioning

confidence: 94%

Research on data association and detection algorithm in point target tracking

He,

Li,

Liu

2023

AOPC 2023: Optical Sensing, Imaging, and Display Technology and Applications; And Biomedical Optics

View full text Add to dashboard Cite

Section: Other Methodsmentioning

confidence: 94%

Research on data association and detection algorithm in point target tracking

He,

Li,

Liu

2023

AOPC 2023: Optical Sensing, Imaging, and Display Technology and Applications; And Biomedical Optics

View full text Add to dashboard Cite

“…In the literature [13], a data association algorithm based on a long short term memory network is proposed, capable of handling multiple target tracking in clutter, learning the probability of association from the noise measurements of the radar and its track, and finally, the association probability can be directly obtained after supervised training, and the results show that the algorithm can effectively improve the accuracy of data association.…”

Section: Improvement Solutionsmentioning

confidence: 99%

Research and development trend of machine learning based track data association algorithm

Zhu

Cai

Cheng

et al. 2023

International Conference on Computer Vision, Application, and Algorithm (CVAA 2022)

View full text Add to dashboard Cite

As the number of the measurement target sensors continues to grow and complicated, and the original track data association algorithm is complexity of the target increases, the process of track association is gradually difficult to meet the actual demand, so it is important to further analyze the multiple target tracking data association algorithm. Machine learning has become a research hotspot in various fields in recent years. In the field of data association, machine learning algorithms can also be used to achieve the processing of track data. Machine learning algorithms can effectively improve the correct association rate compared to traditional methods. Firstly, the basic content of the track data association problem is described; secondly, two common types of research under the multiple target tracking data association problems based on machine learning are analyzed and summarized; after that, the future development trend is predicted given the current research status; finally, the overall research content is summarized.

show abstract

“…The goal of T2TA is to associate tracks from different sensors with the same target. The current T2TA algorithms mainly include three categories: fuzzy mathematical algorithms [22][23][24], artificial intelligence-based algorithms [25,26], statistical mathematical algorithms [27][28][29][30][31][32]. Among them, statistical methods directly use the track state to calculate the similarity to give association results, including maximum likelihood methods [28][29][30][31], belief propagation methods [27,32], etc.…”

Section: Introductionmentioning

confidence: 99%

Multi-Sensor Distributed Fusion based on cross-location for passive tracking

Xu,

Guo,

Kuang

et al. 2024

Preprint

View full text Add to dashboard Cite

To solve the problem of passive multi-sensor tracking in clutter, a distributed fusion algorithm based on trajectory integral distance and pseudo-measurement is proposed. First, a twofold Gaussian mixture model is used to quantize the target measurement signal, and achieve a more accurate estimate of the initial target state. The labelled multi-Bernoulli filter is employed in adjusting the number of tracking targets and getting single-sensor tracks. Subsequently, in the fusion center, a track-to-track association method based on global nearest-neighbour is employed. Trajectory integration distance is utilized to improve association performances. A maximum-likelihood Cross-location track-to-track fusion strategy using pseudo-measurement transformation is proposed to enhance fusion accuracy in passive scenarios. Simulation results verify the performance of the proposed distributed fusion method.

show abstract

DeepDA: LSTM-based Deep Data Association Network for Multi-Targets Tracking in Clutter

Cited by 22 publications

References 31 publications

Research on data association and detection algorithm in point target tracking

Research on data association and detection algorithm in point target tracking

Research and development trend of machine learning based track data association algorithm

Multi-Sensor Distributed Fusion based on cross-location for passive tracking

Contact Info

Product

Resources

About