Multi-Spectral Fusion using Generative Adversarial Networks for UAV Detection of Wild Fires

Kacker, Tanmay; Perrusquía, Adolfo; Guo, Weisi

doi:10.1109/icaiic57133.2023.10067042

Cited by 8 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, processing and analyzing such a volume of data requires the use of modern and effective tools. In this context, machine learning techniques such as extreme gradient boosting [4]- [6], logistic regression [7]- [9], and vanilla convolutional neural networks (vanilla CNN) [10]- [12] are emerging as promising tools that offer advanced solutions for wildfire management: prediction and early detection based on multispectral image analysis. The purpose of this study is to comparatively analyze the effectiveness and applicability of these machine learning methods in fire detection tasks.…”

Section: Introductionmentioning

confidence: 99%

Predicting and detecting fires on multispectral images using machine learning methods

Aitimov,

Kaldarova,

Kassymova

et al. 2024

IJECE

View full text Add to dashboard Cite

In today's world, fire forecasting and early detection play a critical role in preventing disasters and minimizing damage to the environment and human settlements. The main goal of the study is the development and testing of machine learning algorithms for automated detection of the initial stages of fires based on the analysis of multispectral images. Within the framework of this study, the capabilities of three popular machine learning methods: extreme gradient boosting, logistic regression, and vanilla convolutional neural network (vanilla CNN), are considered in the task of processing and interpreting multispectral images to predict and detect fires. XGBoost, as a gradient-boosted decision tree algorithm, provides high processing speed and accuracy, logistic regression stands out for its simplicity and interpretability, while vanilla CNN uses the power of deep learning to analyze spatial and spectral data. The results of the study show that the integration of these methods into monitoring systems can significantly improve the efficiency of early fire detection, as well as help in predicting potential fires.

show abstract

Section: Introductionmentioning

confidence: 99%

Predicting and detecting fires on multispectral images using machine learning methods

Aitimov,

Kaldarova,

Kassymova

et al. 2024

IJECE

View full text Add to dashboard Cite

show abstract

“…The spectrogram [4]- [8] is used as a feature representation in classification tasks. This generates a visual representation that can be exploited by machine learning or deep learning models, e.g, convolutional neural networks (CNNs) to detect spatial patterns associated to the profile [8]- [10], or recurrent neural networks (RNNs) [11], [12] to model timedependencies from consecutive visual frames. Other visual representations that can be further used includes: the range Doppler matrix [13], the cepstrogram [5], [14], or the Cadence velocity diagram (CVD) [14] which can enhance current drone detection and classification systems.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Deep Neural Networks for Drone High Level Intent Classification using Non-Cooperative Radar Data

Fraser,

Perrusquía,

Panagiotakopoulos

et al. 2023

2023 3rd International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME)

Self Cite

View full text Add to dashboard Cite

The proliferation of drones has brought many benefits in different industrial and government sectors due to their low cost and potential applications. Nevertheless, the security and air space can be compromised due to anomalous performances derived to negligence or intentional malicious activities. Thus, identify the hidden intentions of drones' mission profiles is paramount to execute adequate countermeasures. In this paper, an hybrid deep neural network architecture is proposed to classify the high level intent of drones' mission profiles using non-cooperative radar. Radar measurements are created synthetically using open access telemetry data of flight trajectories. The proposed architecture exploits the classification and reconstruction capabilities of deep neural models to classify the drones hidden high-level intent. Several experiments and comparisons are carried out to verify the effectiveness of the proposed approach.

show abstract

“…Here, the approaches are based on classification tasks using image sequences to predict pedestrian trajectories. Deep generative models [21], [22] have been applied to predict the long-term human trajectory conditioned on the long-term objective of the task. In terms of drone intent prediction, genetic algorithms have been adopted to model dangerous drone intents using a support vector regressor (SVR) model of the drone kinematics based on ADS-B flight data [2].…”

Section: Introductionmentioning

confidence: 99%

A Deep Mixture of Experts Network for Drone Trajectory Intent Classification and Prediction using Non-Cooperative Radar Data

Fraser,

Perrusquía,

Panagiotakopoulos

et al. 2023

2023 IEEE Symposium Series on Computational Intelligence (SSCI)

Self Cite

View full text Add to dashboard Cite

The intent prediction of unmanned aerial vehicles (UAVs) also known as drones is a challenging task due to the different mission profiles and tasks that the drone can perform. To alleviate this issue, this paper proposes a deep mixture of experts network to classify and predict drones trajectories measured from non-cooperative radars. Telemetry data of openaccess datasets are converted to simulated radar tracks to generate a pool of heterogeneous trajectories and construct three independent datasets to train, validate, and test the proposed architecture. The network is composed of two main components: i) a deep network that predicts the class associated to the input trajectories and ii) a set of deep experts models that learns the extreme bounds of the trajectories in different future time steps. The proposed approach is tested and compared with different deep models to verify its effectiveness under different flight profiles and time-windows.

show abstract

Multi-Spectral Fusion using Generative Adversarial Networks for UAV Detection of Wild Fires

Cited by 8 publications

References 20 publications

Predicting and detecting fires on multispectral images using machine learning methods

Predicting and detecting fires on multispectral images using machine learning methods

Hybrid Deep Neural Networks for Drone High Level Intent Classification using Non-Cooperative Radar Data

A Deep Mixture of Experts Network for Drone Trajectory Intent Classification and Prediction using Non-Cooperative Radar Data

Contact Info

Product

Resources

About