Images captured under low-light conditions are typically prone to undesirable visual phenomena, particularly color distortion and additive noise, that impede the aesthetics and performance of high-level vision tasks. Therefore, developing an enhancement method is necessary to obtain visually pleasing images. This paper proposes deep multiple networks for low-light image enhancement based on the Retinex model (defined as DMT-Net), wherein each network is configured to perform its own role in the YCbCr color space. More specifically, Decoupled-Net is presented to accurately decouple the luminance channel into reflectance and illumination components using the Retinex model. Denoising-Net is connected to Decoupled-Net to eradicate additive noise in the reflectance component. Boosting-Net enhances the illumination component and reduces halo artifacts. In addition, we propose Chrominance- Net to mitigate the distortion of the Cb and Cr channels. Our study also focuses on having a suitable norm order for the loss function based on kurtosis characteristics, such that each network is properly trained to minimize the residual in the training process. Finally, the experimental results prove that our proposed method outperforms numerous other cutting-edge methods in both qualitative and quantitative terms.

Facial expression recognition is crucial for understanding human emotions and nonverbal communication. With the growing prevalence of facial recognition technology and its various applications, accurate and efficient facial expression recognition has become a significant research area. However, most previous methods have focused on designing unique deep-learning architectures while overlooking the loss function. This study presents a new loss function that allows simultaneous consideration of inter- and intra-class variations to be applied to CNN architecture for facial expression recognition. More concretely, this loss function reduces the intra-class variations by minimizing the distances between the deep features and their corresponding class centers. It also increases the inter-class variations by maximizing the distances between deep features and their non-corresponding class centers, and the distances between different class centers. Numerical results from several benchmark facial expression databases, such as Cohn-Kanade Plus, Oulu-Casia, MMI, and FER2013, are provided to prove the capability of the proposed loss function compared with existing ones.

This letter proposes a novel model for effectively detecting malicious messages in controller area network (CAN) communication, which is widely used in automotive networks. Because in-vehicle networks operate in resource-constrained environments, an intrusion detection system (IDS) must simultaneously provide a low computational load and excellent detection performance. However, existing models are unsuitable for deployment on low-power embedded devices owing to their high computational requirements. This letter presents a low-complexity convolutional neural network (CNN)-based IDS for deployment on embedded edge devices. The proposed model applies CNN operations separately to the CAN ID sequence and the data field of the CAN frame to extract features and concatenate them for feature fusion. Experimental results demonstrate that this approach requires considerably less computational load and provides superior detection performance. Furthermore, the proposed model is deployed on a resource-constrained nRF52840 microcontroller using TensorFlow Lite for Microcontrollers with 20.44 KB flash memory and 26.44 KB RAM without quantization.

Image denoising is regarded as an ill-posed problem in computer vision tasks that removes additive noise from imaging sensors. Recently, several convolution neural network-based image-denoising methods have achieved remarkable advances. However, it is difficult for a simple denoising network to recover aesthetically pleasing images owing to the complexity of image content. Therefore, this study proposes a multi-branch network to improve the performance of the denoising method. First, the proposed network is designed based on a conventional autoencoder to learn multi-level contextual features from input images. Subsequently, we integrate two modules into the network, including the Pyramid Context Module (PCM) and the Residual Bottleneck Attention Module (RBAM), to extract salient information for the training process. More specifically, PCM is applied at the beginning of the network to enlarge the receptive field and successfully address the loss of global information using dilated convolution. Meanwhile, RBAM is inserted into the middle of the encoder and decoder to eliminate degraded features and reduce undesired artifacts. Finally, extensive experimental results prove the superiority of the proposed method over state-of-the-art deep-learning methods in terms of objective and subjective performances.

The controller area network (CAN) protocol, widely used for in-vehicle communication, lacks built-in security features and is inherently vulnerable to various attacks. Numerous attack techniques against CAN have been reported, leading to intrusion detection systems (IDSs) tailored for in-vehicle networks. In this study, we propose a novel lightweight unsupervised IDS for CAN networks, designed for real-time, on-device implementation. The proposed autoencoder model was trained exclusively on normal data. A portion of the attack data was utilized to determine the optimal detection threshold using a Gaussian kernel density estimation function, while the frame count was selected based on error rate analysis. Subsequently, the model was evaluated using four types of attack data that were not seen during training. Notably, the model employs a single threshold across all attack types, enabling detection using a single model. Furthermore, the designed software model was optimized for hardware implementation and validated on an FPGA under a real-time CAN communication environment. When evaluated, the proposed system achieved an average accuracy of 99.2%, precision of 99.2%, recall of 99.1%, and F1-score of 99.2%. Furthermore, compared to existing FPGA-based IDS models, our model reduced the usage of LUTs, flip-flops, and power by average factors of 1/5, 1/6, and 1/11.

The Controller Area Network (CAN), widely used for vehicular communication, is vulnerable to multiple types of cyber-threats. Attackers can inject malicious messages into the CAN bus through various channels, including wireless methods, entertainment systems, and on-board diagnostic ports. Therefore, it is crucial to develop a reliable intrusion detection system (IDS) capable of effectively distinguishing between legitimate and malicious CAN messages. In this paper, we propose a novel IDS architecture aimed at enhancing the cybersecurity of CAN bus systems in vehicles. Various machine learning (ML) models have been widely used to address similar problems; however, although existing ML-based IDS are computationally efficient, they suffer from suboptimal detection performance. To mitigate this shortcoming, our architecture incorporates specially designed rule-based filters that cross-check outputs from the traditional ML-based IDS. These filters scrutinize message ID and payload data to precisely capture the unique characteristics of three distinct types of cyberattacks: DoS attacks, spoofing attacks, and fuzzy attacks. Experimental evidence demonstrates that the proposed architecture leads to a significant improvement in detection performance across all utilized ML models. Specifically, all ML-based IDS achieved an accuracy exceeding 99% for every type of attack. This achievement highlights the robustness and effectiveness of our proposed solution in detecting potential threats.