Video Scene Detection Using Transformer Encoding Linker Network (TELNet)

Tseng, Shu-Ming; Yeh, Zhi-Ting; Wu, Chien-Hung; Chang, Jane W.; Norouzi, Mehdi

doi:10.3390/s23167050

Cited by 4 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Firstly, we add multiple noise sources (Gaussian Noise, Salt and Pepper Noise, Dropout Noise) at the input end of the model to comprehensively consider the feature extraction ability under different noisy conditions. Secondly, we incorporate a self-attention mechanism [45,46] into the hidden layer of the encoder to utilize the correlation between high-dimensional features and use the importance degree to focus the neural network's attention more on important information among various features. Finally, we deepen the autoencoder's layers and use a deep neural network to learn the deep feature representation of high-dimensional features to improve the robustness of the dimensionality reduction model.…”

Section: Feature Dimension Reduction Module Based On Mdsaementioning

confidence: 99%

Machine Learning-Based Intrusion Detection for Rare-Class Network Attacks

Yang,

Gu,

Yan

2023

Electronics

View full text Add to dashboard Cite

Due to the severe imbalance in the quantities of normal samples and attack samples, as well as among different types of attack samples, intrusion detection systems suffer from low detection rates for rare-class attack data. In this paper, we propose a geometric synthetic minority oversampling technique based on the optimized kernel density estimation algorithm. This method can generate diverse rare-class attack data by learning the distribution of rare-class attack data while maintaining similarity with the original sample features. Meanwhile, the balanced data is input to a feature extraction module built upon multiple denoising autoencoders, reducing information redundancy in high-dimensional data and improving the detection performance for unknown attacks. Subsequently, a soft-voting ensemble learning technique is utilized for multi-class anomaly detection on the balanced and dimensionally reduced data. Finally, an intrusion detection system is constructed based on data preprocessing, imbalance handling, feature extraction, and anomaly detection modules. The performance of the system was evaluated using two datasets, NSL-KDD and N-BaIoT, achieving 86.39% and 99.94% multiclassification accuracy, respectively. Through ablation experiments and comparison with the baseline model, it is found that the inherent limitations of a single machine-learning model directly affect the accuracy of the intrusion detection system, while the superiority of the proposed multi-module model in detecting unknown attacks and rare classes of attack traffic is demonstrated.

show abstract

Section: Feature Dimension Reduction Module Based On Mdsaementioning

confidence: 99%

Machine Learning-Based Intrusion Detection for Rare-Class Network Attacks

Yang,

Gu,

Yan

2023

Electronics

View full text Add to dashboard Cite

show abstract

Section: Feature Dimension Reduction Module Based On Mdsaementioning

confidence: 99%

Deep Learning-Based Intrusion Detection for Rare Class Network Attacks

Yang,

Gu,

Yan

2023

Preprint

View full text Add to dashboard Cite

With the continuous development of network technology, complex network systems generate massive unbalanced attack traffic. Due to the severe imbalance in the quantities of normal samples and attack samples, as well as among different types of attack samples, intrusion detection systems suffer from low detection rates for rare class attack data. In this paper, we propose a geometric synthetic minority oversampling technique based on optimized kernel density estimation algorithm. This method can generate diverse rare class attack data by learning the distribution of rare class attack data while maintaining similarity with the original sample features. Meanwhile, the balanced data is input to a feature extraction module built upon multiple denoising autoencoders, reducing information redundancy in high-dimensional data and improving the detection performance for unknown attacks. Subsequently, a soft voting ensemble learning technique is utilized for multi-class anomaly detection on the balanced and dimensionally reduced data. Finally, an intrusion detection system is constructed based on data preprocessing, imbalance handling, feature extraction, and anomaly detection modules, and validated on the NSL-KDD and N-BaIoT datasets. Comparative experiments with baseline models and other state-of-the-art methods demonstrate that the proposed system improves the detection rate of rare class attack data. Furthermore, it achieves a good overall detection rate on the Internet of Things dataset (N-BaIoT), indicating its strong applicability.

show abstract

“…Highlight detection is the process of identifying interesting or important segments within a video [22][23][24][25]. Traditional datasets [26] do not provide personalized highlights as they lack queries related to specific video segments.…”

Section: Introductionmentioning

confidence: 99%

Cross-Modality Interaction-Based Traffic Accident Classification

Oh,

Ban

2024

Applied Sciences

View full text Add to dashboard Cite

Traffic accidents on the road lead to serious personal and material damage. Furthermore, preventing secondary accidents caused by traffic accidents is crucial. As various technologies for detecting traffic accidents in videos using deep learning are being researched, this paper proposes a method to classify accident videos based on a video highlight detection network. To utilize video highlight detection for traffic accident classification, we generate information using the existing traffic accident videos. Moreover, we introduce the Car Crash Highlights Dataset (CCHD). This dataset contains a variety of weather conditions, such as snow, rain, and clear skies, as well as multiple types of traffic accidents. We compare and analyze the performance of various video highlight detection networks in traffic accident detection, thereby presenting an efficient video feature extraction method according to the accident and the optimal video highlight detection network. For the first time, we have applied video highlight detection networks to the task of traffic accident classification. In the task, the most superior video highlight detection network achieves a classification performance of up to 79.26% when using video, audio, and text as inputs, compared to using video and text alone. Moreover, we elaborated the analysis of our approach in the aspects of cross-modality interaction, self-attention and cross-attention, feature extraction, and negative loss.

show abstract

Video Scene Detection Using Transformer Encoding Linker Network (TELNet)

Cited by 4 publications

References 42 publications

Machine Learning-Based Intrusion Detection for Rare-Class Network Attacks

Machine Learning-Based Intrusion Detection for Rare-Class Network Attacks

Deep Learning-Based Intrusion Detection for Rare Class Network Attacks

Cross-Modality Interaction-Based Traffic Accident Classification

Contact Info

Product

Resources

About