Improve Detection Performance Research Articles

TD-Net:tiny defect detection network for industrial products

AbstractThe detection of tiny defects in industrial products is important for improving the quality of industrial products and maintaining production safety. Currently, image-based defect detection methods are ineffective in detecting tiny and variously shaped defects. Therefore, this paper proposes a tiny defect detection network (TD-Net) for industrial products to improve the effectiveness of tiny defect detection. TD-Net improves the overall defect detection effect, especially the detection effect of tiny defects, by solving the problems of downsampling of tiny defects, pre-filtering of conflicting deep and shallow semantic information, and cascading fusion of multi-scale information. Specifically, this paper proposes the Defect Downsampling (DD) module to realize the defect information supplementation during the backbone downsampling process and improve the problem that the stepwise convolution easily misses the detection of tiny defects. Meanwhile, the Semantic Information Interaction Module (SIIM) is proposed, which fuses deep and shallow semantic features, and is designed to interact the fused features with shallow features to optimize the detection of tiny defects. Finally, the Scale Information Fusion Module (SIFM) is proposed to improve the Path Aggregation Network (PANet) for cascading fusion and information focus on different scale information, which enables further improvement of defect detection performance of TD-Net. Extensive experimental results on the NEU–DET data set (76.8$$\%$$ % mAP), the Peking University PCB defect data set (96.2$$\%$$ % mAP) and the GC10-DET data set (71.5$$\%$$ % mAP) show that the proposed TD-Net achieves competitive results compared with SOTA methods with the equivalent parameter quantity.

PeerG: A P2P botnet detection method based on representation learning and graph contrastive learning

P2P botnets are distributed with complex topology and communication behavior, making them harder to detect and remove. Individuals or organizations can effectively detect P2P botnets by analyzing abnormal behaviors in network traffic. Existing works focus on extracting deterministic traffic interaction features, which are highly dependent on statistical features. Moreover, these methods are mainly aimed at generalized botnet detection and lack specificity for detecting P2P botnets. They ignore the topological information of P2P botnets, resulting in unsatisfactory detection accuracy. Among the few existing methods targeting P2P botnets, they consider the topological information but mainly rely on classical graph theory statistical features, such as degree, feature vector centrality, etc. This limits the generalization ability of the detection model. In this paper, we delve into the topological features of P2P botnets from the perspective of complex graph theory. We propose a P2P botnet detection method that combines representation learning and graph contrastive learning, dubbed PeerG. Specifically, we construct the communication graphs based on the flow interaction behavior between components of the P2P botnets. The Line algorithm is employed to embed the nodes of the communication graph into a low-dimensional representation vector space. Subsequently, the graph contrastive learning approach is utilized to optimize the feature extractor, enabling it to capture more representative node features. PeerG serves as a benchmark detection model for identifying P2P botnet nodes. Additionally, we devise two optimized contrastive detection strategies (PeerG-PreG and PeerG-PreF) based on graph-level and feature-level to boost the performance of PeerG. Extensive experiments demonstrate that PeerG brings significant improvements in detection accuracy over state-of-art detection methods. Furthermore, compared with PeerG, the detection strategies PeerG-PreG and PeerG-PreF have further improved detection performance, and achieve the best detection accuracy among multiple filtered P2P botnet types.

Designing A Pdf Malware Detection System Using Machine Learning

This research proposes an approach to build malicious PDF detection system using random forest algorithm, focusing the Evasive-PDFMal2022 dataset which is updated and extended with the addition of new datasets. This dataset includes malicious PDF files from CVE and Exploit-DB, non-malicious PDF files, as well as files from private collections and Technically-oriented PDF Collection. Features were extracted using the PDFID tool, resulting in 29 structural features that formed the basis for the Random Forest classification algorithm. Experiments showed that the model trained with the new dataset provided accuracy equivalent to the Evasive-PDFMal2022 model, at 98%, albeit with a small decrease in recall for the benign class. In addition, this research involved the creation of a website for metadata extraction and malicious PDF detection. Recognition goes to the dataset contributors, tool developers, and dataset providers from NIST and Exploit-DB. Overall, this research successfully increased the representation and diversity of the dataset, provided good model training results, improved detection from 3 malicious PDF variants to 13 variants, and created a practical tool for malicious PDF extraction and detection. Nonetheless, further development may be required to improve detection performance in more complex scenarios

Vegetable disease detection using an improved YOLOv8 algorithm in the greenhouse plant environment

This study introduces YOLOv8n-vegetable, a model designed to address challenges related to imprecise detection of vegetable diseases in greenhouse plant environment using existing network models. The model incorporates several improvements and optimizations to enhance its effectiveness. Firstly, a novel C2fGhost module replaces partial C2f. with GhostConv based on Ghost lightweight convolution, reducing the model’s parameters and improving detection performance. Second, the Occlusion Perception Attention Module (OAM) is integrated into the Neck section to better preserve feature information after fusion, enhancing vegetable disease detection in greenhouse settings. To address challenges associated with detecting small-sized objects and the depletion of semantic knowledge due to varying scales, an additional layer for detecting small-sized objects is included. This layer improves the amalgamation of extensive and basic semantic knowledge, thereby enhancing overall detection accuracy. Finally, the HIoU boundary loss function is introduced, leading to improved convergence speed and regression accuracy. These improvement strategies were validated through experiments using a self-built vegetable disease detection dataset in a greenhouse environment. Multiple experimental comparisons have demonstrated the model's effectiveness, achieving the objectives of improving detection speed while maintaining accuracy and real-time detection capability. According to experimental findings, the enhanced model exhibited a 6.46% rise in mean average precision (mAP) over the original model on the self-built vegetable disease detection dataset under greenhouse conditions. Additionally, the parameter quantity and model size decreased by 0.16G and 0.21 MB, respectively. The proposed model demonstrates significant advancements over the original algorithm and exhibits strong competitiveness when compared with other advanced object detection models. The lightweight and fast detection of vegetable diseases offered by the proposed model presents promising applications in vegetable disease detection tasks.

FastPFM: a multi-scale ship detection algorithm for complex scenes based on SAR images

Synthetic Aperture Radar (SAR) is renowned for its all-weather capabilities, exceptional penetration, and high-resolution imaging, making SAR-based ship detection crucial for maritime surveillance and sea rescue operations. However, various challenges, such as blurred ship contours, complex backgrounds, and uneven scale distribution, can impede detection performance improvement. In this study, we propose FastPFM, a novel ship detection model developed to address these challenges. Firstly, we utilize FasterNet as the backbone network to reduce computational redundancy, enhancing feature extraction efficiency and overall computational performance. Additionally, we employ the Feature Bi-level Routing Transformation model (FBM) to obtain global feature information and enhance focus on target regions. Secondly, the PFM module is engineered to collect multi-scale target information effectively by establishing connections across stages, thereby improving fusion of target features. Thirdly, an extra target feature fusion layer is introduced to enhance small ship detection precision and accommodate multi-scale targets. Finally, comprehensive tests on SSDD and HRSID datasets validate FastPFM's efficacy. Compared to the baseline model YOLOX, FastPFM achieves a 5.5% and 4.4% improvement in detection accuracy, respectively. Furthermore, FastPFM demonstrates comparable or superior performance to other detection algorithms, achieving 92.1% and 83.1% accuracy on AP50, respectively.

Design of Industrial IoT Intrusion Security Detection System Based on LightGBM Feature Algorithm and Multi-layer Perception Network

Continuous improvement of machine learning technology has provided more support for intrusion security detection in the industrial Internet of Things (IoT). The intrusion security detection system based on LightGBM feature algorithm and multi-layer perception network fusion provides more options for improving security detection, further enhancing the effectiveness of industrial IoT intrusion security detection. By comparing the applications of different models in the security detection process, the model constructed based on the LightGBM feature algorithm can achieve higher accuracy and precision in industrial IoT intrusion security detection, as well as higher F1-score and AUC values; A more reasonable detection time also lays the foundation for improving the overall efficiency of industrial IoT intrusion security detection. Therefore, in the field of industrial IoT intrusion security detection, the detection model constructed in this article can provide more support for further improvement and improvement of IoT intrusion security detection performance.

Pulmonary nodule detection in x-ray images by feature augmentation and context aggregation

Recent developments in x-ray image based pulmonary nodule detection have achieved remarkable results. However, existing methods are focused on transferring off-the-shelf coarse-grained classification models and fine-grained detection models rather than developing a dedicated framework optimized for nodule detection. In this paper, we propose PN-DetX, which as we know is the first dedicated pulmonary nodule detection framework. PN-DetX incorporates feature fusion and self-attention into x-ray based pulmonary nodule detection tasks, achieving improved detection performance. Specifically, PN-DetX adopts CSPDarknet backbone to extract features, and utilizes feature augmentation module to fuse features from different levels followed by context aggregation module to aggregate semantic information. To evaluate the efficacy of our method, we collect a LArge-scale Pulmonary NOdule Detection dataset, LAPNOD, comprising 2954 x-ray images along with expert-annotated ground truths. As we know, this is the first large-scale chest x-ray pulmonary nodule detection dataset. Experiments demonstrates that our method outperforms baseline by 3.8% mAP and 5.1% AP 0.5. The generality of our approach is also evaluated on the publicly available dataset NODE21. We aspire for our method to serve as an inspiration for future research in the field of pulmonary nodule detection. The dataset and codes will be made in public.

Vehicle detection and tracking in intelligent transportation systems based deep learning

This research paper focuses on the advancements and optimizations made to fundamental object detection algorithms in vehicle detection. The study explores integrating and reusing CNN (Convolutional Neural Networks) models with other techniques to enhance performance. Three main models, namely Faster R-CNN (Faster Region-based Convolutional Neural Network), Improved SSD (Single Shot Multibox Detector), and YOLOv4 (You Only Look Once v4), are analyzed, showcasing their incremental improvements in accuracy and overall detection performance. However, the increased computational complexity and time demands are trade-offs. The study also presents EnsembleNet, a model combining Faster R-CNN and YOLOv5, which achieves higher average precision values. Another approach involves fusing edge features with CNN models, resulting in faster and more accurate vehicle recognition. The paper predicts future deep learning trends, emphasizing the need for improved hardware capabilities to handle complex models. Integrating deep learning with sensor fusion and edge computing holds promise for intelligent transportation systems.

Sleep apnea detection from single-lead electrocardiogram signals using effective deep-shallow fusion network

Objective. Explore a network architecture that can efficiently perform single-lead electrocardiogram (ECG) sleep apnea (SA) detection by utilizing the beneficial information of extended ECG segments and reducing the impact of their noisy information. Approach. We propose an effective deep-shallow fusion network (EDSFnet). The deeper residual network is used to extract high-level features with stronger semantics and less noise from the original ECG segments. The shallower convolutional neural network is used to extract lower-level features with higher resolution containing more detailed neighborhood information from the extended ECG segments. These two types of features are then fused using Effective Channel Attention, implementing automatic weight assignment to take advantage of their complementary nature. Main results. The performance of EDSFnet is evaluated on the Apnea-ECG dataset and the FAH-ECG dataset. In the Apnea-ECG dataset with 35 subjects as the training set and 35 subjects as the test set, the accuracy of EDSFnet was 92.6% and 100% for per-segment and per-recording test, respectively. In the FAH-ECG dataset with 348 subjects as the training set and 88 subjects as the test set, the accuracy of EDSFnet was 89.0% and 93.2% for per-segment and per-recording test, respectively. EDSFnet has achieved state-of-the-art results in both experiments using the publicly available Apnea-ECG dataset and subject-independent experiments using the FAH-ECG clinical dataset. Significance. The success of EDSFnet in handling SA detection underlines its robustness and adaptability. By achieving superior results across different datasets, EDSFnet offers promise in advancing the cost-effective and efficient detection of SA through single-lead ECG, reducing the burden on patients and healthcare systems alike.

Optimized Cluster-Based Cooperative Spectrum Sensing Over Weibull, Nakagami, Rician, and Rayleigh Fading Channels

This paper reports cluster-based cooperative spectrum sensing (CB-CSS) over Weibull, Nakagami, Rician, and Rayleigh fading channels in cognitive radio networks (CRNs). The performance assessment of centralized cooperative spectrum sensing (CCSS) and CB-CSS with different fusion rules over the Weibull fading channel is presented. The goal is to improve detection performance in a practical and efficient manner for CB-CSS. For this purpose, an expression for the optimal inter-fusion rule threshold (fusion rule between cluster head (CH) and fusion center (FC)) is proposed to minimize total error probability (TEP) over Weibull for CB-CSS. Simulations are conducted using analytical equations for missed detection and false alarm probability. The obtained expression for optimal inter-fusion rule threshold which minimizes TEP is verified by the results. The proposed model is validated with other fading channels such as Nakagami, Rician, and Rayleigh to show the superiority of the approach. It is observed that for a given range of detection threshold and for 4 × 6 cluster size (4 clusters and 6 CR users inside each cluster), the lowest total error probability is achieved over the Weibull fading channel. Minimum TEP over the Weibull fading channel for given network parameters is TEP = 1.656 × 10 − 6 when intra- and inter-fusion rules are l = 1 (OR fusion) and c = 3 . The least number of clusters and users essential to achieving target error bound over different fading channels are also determined.

Towards Fully Autonomous UAV: Damaged Building-Opening Detection for Outdoor-Indoor Transition in Urban Search and Rescue

Autonomous unmanned aerial vehicle (UAV) technology is a promising technology for minimizing human involvement in dangerous activities like urban search and rescue missions (USAR), both in indoor and outdoor. Automated navigation from outdoor to indoor environments is not trivial, as it encompasses the ability of a UAV to automatically map and locate the openings in a damaged building. This study focuses on developing a deep learning model for the detection of damaged building openings in real time. A novel damaged building-opening dataset containing images and mask annotations, as well as a comparison between single and multi-task learning-based detectors are given. The deep learning-based detector used in this study is based on YOLOv5. First, this study compared the different versions of YOLOv5 (i.e., small, medium, and large) capacity to perform damaged building-opening detections. Second, a multitask learning YOLOv5 was trained on the same dataset and compared with the single-task detector. The multitask learning (MTL) was developed based on the YOLOv5 object detection architecture, adding a segmentation branch jointly with the detection head. This study found that the MTL-based YOLOv5 can improve detection performance by combining detection and segmentation losses. The YOLOv5s-MTL trained on the damaged building-opening dataset obtained 0.648 mAP, an increase of 0.167 from the single-task-based network, while its inference speed was 73 frames per second on the tested platform.

Relation-aware graph convolutional network for waste battery inspection based on X-ray images

With the popularization of various electronic devices, the demand for batteries is rapidly increasing. Batteries bring convenience to our daily lives, but how to effectively achieve battery recycling is a challenge. In fact, improper disposal can cause serious environmental pollution and resource waste. Currently, it is inefficient and time-consuming to manually locate batteries from waste electronic devices by employing a large number of professional workers. Therefore, we propose an efficient solution based on deep learning and X-ray images, which can automatically detect batteries without opening waste electronic devices. Specifically, we construct a High-quality X-ray Battery Inspection dataset called HiXray-BI, which not only contains 6000 X-ray images collected from the existing dataset, but also provides 23,994 battery instances. In addition, we propose a novel deep learning model called Relation-Aware Graph Convolutional Network (RA-GCN) for X-ray waste battery inspection, which introduces a novel Multi-scale Feature Relation Module (MFRM) and Semantic Feature Relation Module (SFRM) to improve detection performance. Experimental results on the HiXray-BI dataset show that our method compares favorably with state-of-the-art methods. Our solution provides new insights into recycling waste batteries, which is beneficial for environmental protection and sustainable development.

Anomaly Detection on the Edge Using Smart Cameras under Low-Light Conditions

The number of cameras utilised in smart city domains is increasingly prominent and notable for monitoring outdoor urban and rural areas such as farms and forests to deter thefts of farming machinery and livestock, as well as monitoring workers to guarantee their safety. However, anomaly detection tasks become much more challenging in environments with low-light conditions. Consequently, achieving efficient outcomes in recognising surrounding behaviours and events becomes difficult. Therefore, this research has developed a technique to enhance images captured in poor visibility. This enhancement aims to boost object detection accuracy and mitigate false positive detections. The proposed technique consists of several stages. In the first stage, features are extracted from input images. Subsequently, a classifier assigns a unique label to indicate the optimum model among multi-enhancement networks. In addition, it can distinguish scenes captured with sufficient light from low-light ones. Finally, a detection algorithm is applied to identify objects. Each task was implemented on a separate IoT-edge device, improving detection performance on the ExDark database with a nearly one-second response time across all stages.

SIANet: 3D object detection with structural information augment network

Abstract3D object detection technology from point clouds has been widely applied in the field of automatic driving in recent years. In practical applications, the shape point clouds of some objects are incomplete due to occlusion or far distance, which means they suffer from insufficient structural information. This greatly affects the detection performance. To address this challenge, the authors design a Structural Information Augment (SIA) Network for 3D object detection, named SIANet. Specifically, the authors design a SIA module to reconstruct the complete shapes of objects within proposals for enhancing their geometric features, which are further fused into the spatial feature of the object for box refinement to predict accurate detection boxes. Besides, the authors construct a novel Unet‐liked Context‐enhanced Transformer backbone network, which stacks Context‐enhanced Transformer modules and an upsampling branch to capture contextual information efficiently and generate high‐quality proposals for the SIA module. Extensive experiments show that the authors’ well‐designed SIANet can effectively improve detection performance, especially surpassing the baseline network by 1.04% mean Average Precision (mAP) gain in the KITTI dataset and 0.75% LEVEL_2 mAP gain in the Waymo dataset.

Securing IIoT applications in 6G and beyond using adaptive ensemble learning and zero-touch multi-resource provisioning

The advent of Industry 5.0, driven by cutting-edge 6G technologies such as immersive cloud eXtended Reality (XR), Autonomous vehicles, holographic communication, and Digital Twin (DT), is set to trigger a substantial upswing in the deployment of the Industrial Internet of Things (IIoT) devices interconnected within networks. This expansion of IIoT devices will create a wider attack surface and increase the risk of data breaches, privacy violations, and system disruptions. Therefore, it is essential to design innovative mechanisms to ensure the reliability and security of these advanced 6G applications, as well as the IIoT devices that support them. In this context, we propose a novel Software-defined Networking (SDN)-based Ensemble Learning (EL) Framework for Secure IIoT applications in 6G and beyond, called AdaptSDN. The proposed framework leverages SDN technology to dynamically allocate network resources and deploy security measures on demand. It also leverages EL techniques to improve the intrusion detection system’s accuracy. By isolating IIoT devices into network slices, the framework limits the impact of attacks and reduces the potential for cascading failures. Digital twins are used for creating a virtual replica of the IIoT network, allowing for a real-time security threat detection. In particular, AdaptSDN includes three main modules: (1) A novel digital Twin (DT)-enabled data gathering and selection of informative features module to achieve two main objectives: reducing computational complexity and improving detection performance; (2) An SDN-based lightweight adaptive boosting module that uses an advanced boosting EL techniques to dynamically adjust weights and to effectively identify and respond to IIoT attacks in real-time; and (3) A zero-touch resources provisioning module that employs a non-cooperative game theory approach. This approach allows for automatically provisioning various resources in a network to efficiently mitigate network attacks; it enables SDN nodes to obtain the required virtual resources, i.e., storage, computing, and bandwidth, from the main orchestrator with respect to IIoT attack type. We carried out comprehensive experiments to assess the effectiveness of our proposed framework in detecting real-world IIoT attacks. The numerical results confirm that AdaptSDN has the potential to enable secure and reliable IIoT applications in 6G and beyond, meeting the stringent service requirements of the new emerging applications.

Comparative Evaluation of Color Correction as Image Preprocessing for Olive Identification under Natural Light Using Cell Phones

Integrating deep learning for crop monitoring presents opportunities and challenges, particularly in object detection under varying environmental conditions. This study investigates the efficacy of image preprocessing methods for olive identification using mobile cameras under natural light. The research is grounded in the broader context of enhancing object detection accuracy in variable lighting, which is crucial for practical applications in precision agriculture. The study primarily employs the YOLOv7 object detection model and compares various color correction techniques, including histogram equalization (HE), adaptive histogram equalization (AHE), and color correction using the ColorChecker. Additionally, the research examines the role of data augmentation methods, such as image and bounding box rotation, in conjunction with these preprocessing techniques. The findings reveal that while all preprocessing methods improve detection performance compared to non-processed images, AHE is particularly effective in dealing with natural lighting variability. The study also demonstrates that image rotation augmentation consistently enhances model accuracy across different preprocessing methods. These results contribute significantly to agricultural technology, highlighting the importance of tailored image preprocessing in object detection models. The conclusions drawn from this research offer valuable insights for optimizing deep learning applications in agriculture, particularly in scenarios with inconsistent environmental conditions.

Automatic detection of building surface cracks using UAV and deep learning‐combined approach

AbstractConcrete cracking is one of the most significant damage types in reinforced concrete structures due to its potential to adversely affect durability, safety, and serviceability and even reduce the bearing capacity during operation. Thus, damage inspection of damage caused by concrete cracking is important for management, maintenance, and structural assessment for both damaged and undamaged existing buildings but with concrete cracking after a long time of use that needs reconstruction or renovation. This study provides an improved building damage inspection approach by applying Unmanned Aerial Vehicles (UAVs) and state‐of‐the‐art deep learning algorithms to detect concrete cracks on building surfaces. Two distinct architectures for Convolutional Neural Networks (CNNs), namely ResNet50 and YOLOv8 based on classification, and object detection approaches to create a total of 11 models are established, trained, and compared. The classification models attained accuracy levels exceeding 99%, whereas the object detection models achieved approximately 85%. All models effectively identified and accurately located concrete cracks on building surfaces. Besides, the CNN models' capacity to detect cracks is influenced by a variety of model hyperparameters, encompassing factors such as model architecture, the number of network layers, different training dataset sizes, and the quantity of trainable parameters necessary to learn the specific features of detection targets during the training process. The results of this study ultimately demonstrate that the proposed approach can yield accurate detection results and holds high potential for application in crack inspection to advance automatic damage inspection in building structures to a greater extent. In addition, it is important to note that a universal rule cannot be established rule as a larger and more complex model, or an increased number of trainable parameters, necessarily leads to improved detection performance. Models that are trained from scratch using local datasets might not necessarily result in enhanced performance in comparison to the improvements gained through fine‐tuning via transfer learning. Therefore, an appropriate training type, dataset size, task complexity, computational resources, and time demands to achieve a balance between accuracy and efficiency should be considered for specific application scenarios.

Urine: A Pitfall for Molecular Detection of Toscana Virus? An Analytical Proof-of-Concept Study.

Toscana virus (TOSV), a sandfly-borne virus, is an important etiological agent in human acute meningitis and meningoencephalitis in the Mediterranean area during the summer. However, the actual number of TOSV infections is underestimated. Laboratory confirmation is necessary because TOSV infection has overlapping clinical features with other neuro-invasive viral infections. Nowadays, the reference test for direct diagnosis in the acute phase of TOSV infection is the PCR based method for detecting TOSV in cerebrospinal fluid and/or plasma, serum, or blood. Although poorly employed, urine is another helpful biological matrix for TOSV detection. Urine is a matrix rich in PCR inhibitors that affect PCR efficiency; consequently, false negatives could be generated. To investigate the potential effect of urine PCR inhibitors on TOSV detection, we compared undiluted and diluted urine using 10-fold series of spiked TOSV. The results showed a significant improvement in TOSV detection performance in diluted urine (1 TCID50 vs. 1 × 104 TCID50 limit of detection and 101.35% vs. 129.62% efficiency, respectively, in diluted and undiluted urine). In conclusion, our data provide preliminary important insights into the use of diluted urine to limit the impact of the inhibitory effects of urine on the detection of TOSV in RT-PCR-based approaches.

Significance of relative phase features for shouted and normal speech classification

Shouted and normal speech classification plays an important role in many speech-related applications. The existing works are often based on magnitude-based features and ignore phase-based features, which are directly related to magnitude information. In this paper, the importance of phase-based features is explored for the detection of shouted speech. The novel contributions of this work are as follows. (1) Three phase-based features, namely, relative phase (RP), linear prediction analysis estimated speech-based RP (LPAES-RP) and linear prediction residual-based RP (LPR-RP) features, are explored for shouted and normal speech classification. (2) We propose a new RP feature, called the glottal source-based RP (GRP) feature. The main idea of the proposed GRP feature is to exploit the difference between RP and LPAES-RP features to detect shouted speech. (3) A score combination of phase- and magnitude-based features is also employed to further improve the classification performance. The proposed feature and combination are evaluated using the shouted normal electroglottograph speech (SNE-Speech) corpus. The experimental findings show that the RP, LPAES-RP, and LPR-RP features provide promising results for the detection of shouted speech. We also find that the proposed GRP feature can provide better results than those of the standard mel-frequency cepstral coefficient (MFCC) feature. Moreover, compared to using individual features, the score combination of the MFCC and RP/LPAES-RP/LPR-RP/GRP features yields an improved detection performance. Performance analysis under noisy environments shows that the score combination of the MFCC and the RP/LPAES-RP/LPR-RP features gives more robust classification. These outcomes show the importance of RP features in distinguishing shouted speech from normal speech.

An object detection algorithm based on infrared-visible dual modal feature fusion

Dual modal image pairs can provide complementary feature information and overcome the limitations of single modal object detection algorithms, thus improving detection performance. In order to fully utilize the features of different modalities, this paper proposes a dual-branch fusion detection network that can simultaneously input both infrared and visible modalities for object detection. The method is based on YOLOv5-s, and two attention modules are proposed according to the characteristics of different modalities to enhance the expression capability of infrared and visible image feature information respectively; At the same time, a dual modal fusion module is designed for cross-modal information complementarity by fusing features from corresponding scales of both modalities; Finally, a feature enhancement module is proposed to improve the multi-scale feature fusion capability. The algorithm is validated on the KAIST, FLIR, and GIR datasets and compared with classic single modal and dual modal detection algorithms. Experimental results show that, compared with the baseline algorithm YOLOv5-s alone detecting visible and infrared images separately, the proposed algorithm improves the detection accuracy by 17.6% and 5.9% on the KAIST dataset respectively, and 19.1% and 13% on the FLIR dataset respectively. The proposed algorithm also has a significant advantage in detection accuracy on the self-built GIR dataset.