Scene Characteristics Research Articles

Identification of Shipborne VHF Radio Based on Deep Learning with Feature Extraction

In the feature identification of maritime VHF radio communication signals, shipborne VHF communication technology follows the same international technical standards formulated by IMO, uses analog communication technology and uses the same communication channel in the same area, and cannot effectively achieve signal feature identification by adding feature elements in the process of signal modulation. How to effectively identify the ship using VHF radio has always been a technical difficulty in the field of ship perception. In this paper, based on the convolutional neural network, combined with the feasibility of CAM feature extraction and BiLSTM feature extraction in non-cooperative signal recognition, a deep learning recognition model of shipborne VHF radio communication signals is established, and the deep learning approach is employed to discern the features of VHF signals, thereby accomplishing the identification and classification of transmitting VHF radio stations. Several experiments are designed according to the characteristics of ship communication scenes at sea. The experimental data show that the method proposed in this paper can provide a new feasible path for ship target perception in terms of radio signal characteristics and identification.

Characterization of fire investigators’ polyaromatic hydrocarbon exposures using silicone wristbands

BackgroundExposures to polyaromatic hydrocarbons (PAHs) contribute to cancer in the fire service. Fire investigators are involved in evaluations of post-fire scenes. In the US, it is estimated that there are up to 9000 fire investigators, compared to approximately 1.1 million total firefighting personnel. This exploratory study contributes initial evidence of PAH exposures sustained by this understudied group using worn silicone passive samplers. ObjectivesEvaluate PAH exposures sustained by fire investigators at post-fire scenes using worn silicone passive samplers. Assess explanatory factors and health risks of PAH exposure at post-fire scenes. MethodsAs part of a cross-sectional study design, silicone wristbands were distributed to 16 North Carolina fire investigators, including eight public, seven private, and one public and private. Wristbands were worn during 46 post-fire scene investigations. Fire investigators completed pre- and post-surveys providing sociodemographic, occupational, and post-fire scene characteristics. Solvent extracts from wristbands were analyzed via gas chromatography-mass spectrometry (GC-MS). Results were used to estimate vapor-phase PAH concentration in the air at post-fire scenes. ResultsFire investigations lasted an average of 148 minutes, standard deviation ± 93 minutes. A significant positive correlation (r=0.455, p<.001) was found between investigation duration and PAH concentrations on wristbands. Significantly greater time-normalized PAH exposures (p=0.039) were observed for investigations of newer post-fire scenes compared to older post-fire scenes. Regulatory airborne PAH exposure limits were exceeded in six investigations, based on exposure to estimated vapor-phase PAH concentrations in the air at post-fire scenes. DiscussionHigher levels of off-gassing and suspended particulates at younger post-fire scenes may explain greater PAH exposure. Weaker correlations are found between wristband PAH concentration and investigation duration at older post-fire scenes, suggesting reduction of off-gassing PAHs over time. Exceedances of regulatory PAH limits indicate a need for protection against vapor-phase contaminants, especially at more recent post-fire scenes.

AM-UNet: Field Ridge Segmentation of Paddy Field Images Based on an Improved MultiResUNet Network

In order to solve the problem of image boundary segmentation caused by the irregularity of paddy fields in southern China, a high-precision segmentation method based on the improved MultiResUNet model for paddy field mapping is proposed, combining the characteristics of paddy field scenes. We introduce the attention gate (AG) mechanism at the end of the encoder–decoder skip connections in the MultiResUNet model to generate the weights and highlight the response of the field ridge area, add an atrous spatial pyramid pooling (ASPP) module after the end of the encoder down-sampling, use an appropriate combination of expansion rates to improve the identification of small-scale edge details, use 1 × 1 convolution to improve the range of the sensory field after bilinear interpolation to increase the segmentation accuracy, and, thus, construct the AM-UNet paddy field ridge segmentation model. The experimental results show that the IoU, precision, and F1 value of the AM-UNet model are 88.74%, 93.45%, and 93.95%, respectively, and that inference time for a single image is 168ms, enabling accurate and real-time segmentation of field ridges in a complex paddy field environment. Thus, the AM-UNet model can provide technical support for the development of vision-based automatic navigation systems for agricultural machines.

Style-Enhanced Transformer for Image Captioning in Construction Scenes.

Image captioning is important for improving the intelligence of construction projects and assisting managers in mastering construction site activities. However, there are few image-captioning models for construction scenes at present, and the existing methods do not perform well in complex construction scenes. According to the characteristics of construction scenes, we label a text description dataset based on the MOCS dataset and propose a style-enhanced Transformer for image captioning in construction scenes, simply called SETCAP. Specifically, we extract the grid features using the Swin Transformer. Then, to enhance the style information, we not only use the grid features as the initial detail semantic features but also extract style information by style encoder. In addition, in the decoder, we integrate the style information into the text features. The interaction between the image semantic information and the text features is carried out to generate content-appropriate sentences word by word. Finally, we add the sentence style loss into the total loss function to make the style of generated sentences closer to the training set. The experimental results show that the proposed method achieves encouraging results on both the MSCOCO and the MOCS datasets. In particular, SETCAP outperforms state-of-the-art methods by 4.2% CIDEr scores on the MOCS dataset and 3.9% CIDEr scores on the MSCOCO dataset, respectively.

Analysis and modeling of radio propagation fading characteristics in tunnel construction

With the development of underground rail transit, the concept of intelligent tunnel construction has been proposed and promoted. High-quality networking during tunnel construction is a prerequisite for this, making it highly urgent to establish networking during tunnel construction. When studying intra-tunnel networking, it is necessary to consider the propagation characteristics of radio waves in the tunnel. In this paper, according to the actual needs of engineering in tunnel construction and the characteristics of tunnel scenes, an improved ray tracing method is proposed, which considers the type and installation position of antennas, transceiver frequency band and power in channel modeling, and proposes a field strength calculation method under different coordinate systems according to the characteristics of straight and curved segments during tunnel construction. In addition, the propagation characteristics of radio waves in dynamic tunnel construction scenarios are quantitatively analyzed. In this paper, by establishing antenna diagram, two-dimensional and three-dimensional models of tunnels, the computer simulation method is applied to compare with the improved algorithm, and the results have good consistency, in addition, the improved algorithm does not require a lot of modeling work in the early stage, and has high applicability and portability. Not only that, this paper also makes actual measurements of the subway under construction in Zhengzhou, China in different scenarios, and verifies the effectiveness of the method.

Predicting Typhoon Flood in Macau Using Dynamic Gaussian Bayesian Network and Surface Confluence Analysis

A typhoon passing through or making landfall in a coastal city may result in seawater intrusion and continuous rainfall, which may cause urban flooding. The urban flood disaster caused by a typhoon is a dynamic process that changes over time, and a dynamic Gaussian Bayesian network (DGBN) is used to model the time series events in this paper. The scene data generated by each typhoon are different, which means that each typhoon has different characteristics. This paper establishes multiple DGBNs based on the historical data of Macau flooding caused by multiple typhoons, and similar analysis is made between the scene data related to the current flooding to be predicted and the scene data of historical flooding. The DGBN most similar to the scene characteristics of the current flooding is selected as the predicting network of the current flooding. According to the topography, the influence of the surface confluence is considered, and the Manning formula analysis method is proposed. The Manning formula is combined with the DGBN to obtain the final prediction model, DGBN-m, which takes into account the effects of time series and non-time-series factors. The flooding data provided by the Macau Meteorological Bureau are used to carry out experiments, and it is proved that the proposed model can predict the flooding depth well in a specific area of Macau under the condition of a small amount of data and that the best predicting accuracy can reach 84%. Finally, generalization analysis is performed to further confirm the validity of the proposed model.

Multi-frame network feature fusion model and self-attention mechanism for vehicle lane line detection

The traditional lane detection networks mainly use independent single frame images to extract features first and then detect them, which cannot deal with the scene with complex background well. Therefore, this paper proposes a lane parallel detection network based on multi-frame network feature fusion model and self-attention mechanism according to the scene characteristics that vehicles can obtain continuous images during normal driving. Firstly, a parallel feature extraction structure is designed. On the one hand, a single frame network with high precision is used to extract the features of the current frame. On the other hand, a lightweight multi-frame network is designed to extract features of low-resolution multi-frame temporal images. And the recurrent neural network module is used to fuse the extracted temporal features and obtain multi-frame features. Self-attention mechanism can effectively capture the relevant information of internal features. Then the fusion module of single frame feature, multi-frame feature and self-attention feature is designed. The feature map of lane line is output by up-sampling network. The experimental results show that the network in this paper has significant improvement in both objective detection accuracy and subjective effect compared with other methods.

MONOCULAR DEPTH ESTIMATION FOR NIGHT-TIME IMAGES

Abstract. Depth estimation plays a pivotal role in numerous computer vision applications. However, depth estimation networks trained exclusively on daytime images tend to yield poor performance when applied to nighttime scenarios due to domain differences and variations in scene characteristics. In order to address this limitation, we conducted experiments involving the creation of a synthetic nighttime dataset by employing image translation techniques through a generative network. Subsequently, we utilized the generated images to fine-tune the depth estimation network, aiming to investigate the potential for enhancing task performance using generated data. We evaluated our approach by testing with the generated data, and we observed a noticeable improvement in the depth estimation task both before and after fine-tuning. Consequently, our approach yields results that are comparable to those achieved by networks specifically designed for daytime prediction. These findings highlight the effectiveness of utilizing synthetic data to enhance the performance of depth estimation tasks, particularly in nighttime settings.

Continuous burned area monitoring using bi-temporal spectral index time series analysis

Extreme wildfires are a major agent of disturbance in present-day forest ecosystems and, have serious impacts on society in terms of human casualties and economic losses. Prioritizing and implementing relief and recovery measures after fires require explicit spatial information on the disaster extent. Information for detecting and monitoring ecosystem disturbance due to wildfires is obtainable using time series analysis of remote sensing data. However, the accuracy of this information depends on many factors, such as the input spectral features and the scene characteristics. In this study, we appraise the use of three spectral indices for continuous burned area monitoring in mainland Greece using the Breaks For Additive Seasonal and Trend (BFAST) Monitor unsupervised algorithm. In addition to the commonly employed Normalized Burn Ratio (NBR) we evaluated the use of differenced Normalized Burn Ratio (dNBR) and Relativized Burn Ratio (RBR) generated from a 16-day Landsat-8 time series. The spatiotemporal Overall Accuracy (OA) is assessed using Copernicus Emergency Mapping Service (EMS) grading products and an independent visual interpretation procedure. The latter approach suggests that the using the dNBR (OA = 87.0 %) and RBR (OA = 86.9 %) indices is more efficient than using the NBR time series (OA = 66.4 %). Copernicus EMS validation data show that the RBR-based approach attains a higher OA (80.5 %) -higher than either dNBR (OA = 75.0 %) or NBR (OA = 72.1 %). Evaluating the temporal agreement with Copernicus EMS products, bi-temporal indices achieved mean time lags of 0.023 (dNBR), 0.019 (RBR) and −0.015 (NBR) years. Overall, RBR usage provides the most reliable input for consistent continuous burned area monitoring.

Diffusion Convolution Neural Network-Based Multiview Gesture Recognition for Athletes in Dynamic Scenes

This paper focuses on deep vision sensing-assisted gesture recognition for athletes in dynamic scenes. Although many research attention had been devoted to this field in recent years, most of existing works failed to fully take characteristics of dynamic scenes into consideration. To deal with this challenge, this paper proposes a diffusion convolution neural network-based multiview gesture recognition approach in dynamic scenes. For one thing, the dynamic spatiotemporal slice position selection based on the body mask heatmap is adopted to calculate positions of horizontal and vertical slices. Thus, the dynamic selection of slice positions in two directions can be realized, and then the extraction of bi-directional spatiotemporal slice images can be completed. For another, action sequences through the 3D residual neural network are learned, and the spatiotemporal information among frames are mined through recurrent networks. Through their combination, a multi-view gesture recognition approach for athletes is constructed. In the experiments, two standard datasets UCF101 and HMDB51 are utilized to establish simulation environment. The proposed method can reach the accuracy beyond 95% on the two datasets. Compared with several typical recognition methods, the proposed method shows higher accuracy.

Aural based scene understanding for sonar applications

Acoustic returns scattered from either surfaces or objects are typically interpreted using image processing techniques such as beamforming and computer vision models such as convolutional neural networks. However, image representations can be burdensome to generate, creating delayed responses in real-time scenarios, and difficult to parse from the scene-understanding perspective. This work investigates direct exploitation of the information content from streaming, near-raw sonar time series data for the purpose of scene characterization. Summary statistics drawn from the sonar performance estimation community and communities that study aural-based perception are used to identify scene characteristics such as flat sand, ripples, gravel beds, or sea grass, in side-looking synthetic aperture sonar (SAS) data. An empirical multi-class classification study is presented that explores the performance of seafloor characterization methods on time series versus imagery.

Locally Adaptive Channel Attention-Based Spatial–Spectral Neural Network for Image Deblurring

Recently, due to the rapid development of deep neural networks in the field of computer vision, many studies have been conducted in the field of image deblurring. However, previous methods cannot attain satisfactory deblurring performance in object boundaries and local regions containing image details in restored results. This paper proposes a new method that uses a locally adaptive channel attention module for a spectral–spatial network to resolve the problem of single-image deblurring. Unlike existing methods, our proposed method consists of a spectral restorer and spatial restorer that adopt a locally adaptive attention mechanism in the spectral–spatial domains. In addition, unlike a conventional spectral–spatial network that only considers the magnitude of the frequency coefficients, the proposed method uses both the magnitude and phase of the frequency coefficients in the training stage. Our locally adaptive channel attention spectral–spatial network can focus on informative channels that are closely related to blur artifacts. Specifically, the spatial restorer, which guides the intensity of the blur image to fit the ground truth by exploring the interdependencies of feature channels, can efficiently restore scene characteristics while the spectral restorer, which guides the magnitude and phase of the blur image’s frequency coefficients to fit those of the ground truth by exploring the interdependencies of feature channels, fine-tunes the details of structures. The experimental results show that the proposed method outperformed the deblurring results compared to benchmark methods in terms of both qualitative evaluation and quantitative metrics.

Using Social Media and Multi-Source Geospatial Data for Quantifying and Understanding Visitor’s Preferences in Rural Forest Scenes: A Case Study from Nanjing

Rapid urbanization has made urban forest scenes scarce resources, leading to a surge in the demand for high-quality rural forest scenes as alternative outdoor recreation spaces. Previous studies mainly applied survey methods, focusing on visitors’ feedback for different types of scenes from the perspective of visual quality evaluation. Nevertheless, the explanations of the relationships between various factors of scenes and visitors’ preferences are relatively superficial. This study sought to explore the distribution and characteristics of preferred rural forest scenes based on visitor reviews from social media, and using Geodetector, a geospatial statistics tool, to quantitatively analyzed the reasons for visitors’ preferences in terms of factors obtained from multi-source geospatial data. The findings are that (1) visitors are already satisfied with the natural environment but expect scenes that reflect the culture of tea; (2) spatial factor has a more robust interpretation of visitors’ preference, and although the Normalized Difference Vegetation Index (NDVI) and non-consumption indicators barely explain visitors’ preference solely when each of them is combined with other indicators, they can produce non-linear enhancement effects. Consequently, this study synthesizes visitors’ feedback and factors in rural forest scenes to understand visitors’ preferences, thus providing insights into human-centered planning.

Naked body disposal: an indicator of the type of sexual homicide

PurposeThe purpose of this study is to investigate whether sexual homicide offenders (SHO) who dispose of the victim’s body naked present with particular crime scene characteristics.Design/methodology/approachThis study aims to answer this question through the use of a sequential logistic regression to test the individual effects of each set of crime scene variables against the manner of disposal using a sample of 662 solved cases of extrafamilial sexual homicide from an international database.FindingsResults demonstrated that the modus operandi behaviors of sexual penetration, asphyxiation, dismemberment and overkill were significantly associated with the body being disposed of naked. In addition, removing or destroying evidence from the scene was also significantly associated with a naked victim. In contrast, the body was more likely to be dumped clothed if the contact scene was deserted and the victim was a stranger. These results suggest that SHOs who dispose of the body naked are more in line with the sadistic sexual murderer, while clothed victims are often disposed of by angry offenders.Originality/valueTo the best of the authors’ knowledge, this is the first study to examine the particular manner of disposing the victim’s body naked in cases of sexual homicide.

Fast 3D Reconstruction of UAV Images Based on Neural Radiance Field

Traditional methods for 3D reconstruction of unmanned aerial vehicle (UAV) images often rely on classical multi-view 3D reconstruction techniques. This classical approach involves a sequential process encompassing feature extraction, matching, depth fusion, point cloud integration, and mesh creation. However, these steps, particularly those that feature extraction and matching, are intricate and time-consuming. Furthermore, as the number of steps increases, a corresponding amplification of cumulative error occurs, leading to its continual augmentation. Additionally, these methods typically utilize explicit representation, which can result in issues such as model discontinuity and missing data during the reconstruction process. To effectively address the challenges associated with heightened temporal expenditures, the absence of key elements, and the fragmented models inherent in three-dimensional reconstruction using Unmanned Aerial Vehicle (UAV) imagery, an alternative approach is introduced—the neural radiance field. This novel method leverages neural networks to intricately fit spatial information within the scene, thereby streamlining the reconstruction steps and rectifying model deficiencies. The neural radiance field method employs a fully connected neural network to meticulously model object surfaces and directly generate the 3D object model. This methodology simplifies the intricacies found in conventional 3D reconstruction processes. Implicitly encapsulating scene characteristics, the neural radiance field allows for iterative refinement of neural network parameters via the utilization of volume rendering techniques. Experimental results substantiate the efficacy of this approach, demonstrating its ability to complete scene reconstruction within a mere 5 min timeframe, thereby reducing reconstruction time by 90% while markedly enhancing reconstruction quality.

Assessing Precision and Dependability of Reconstructed Three-Dimensional Modeling for Vehicles at Crash Scenes using Unmanned Aircraft System

This study focuses on the accuracy assessment of 3D reconstructions of crime scenes using Unmanned Aircraft Systems (UAS) and Terrestrial Laser Scanners (TLS) data for forensic crash investigation. Forensic crash investigation involves meticulously analyzing physical evidence, vehicles, and human factors in road collisions to determine the sequence of events. Preserving the original state of the crash scene before cleaning is essential for accurate forensic analysis. However, this preservation process can disrupt normal activities and demand considerable time. Geomatic technology, specifically UAS or drones, offers a potential solution for efficient and precise forensic mapping. The application of UAS technology enables swift data collection, leading to cost savings, enhanced safety, and data utilization. This study aims to assess the suitability of UAS techniques for forensic mapping, encompassing both relative and absolute accuracy. This research uses a UAS to rapidly and comprehensively capture evidence from a simulated crash site using predefined flight paths. The acquired image data is then processed utilizing Agisoft Metashape software, generating a detailed 3D model of the crash scene. This model can be enriched with annotations, measurements, and pertinent information. A comparative analysis is performed by preparing a table that contrasts the absolute and relative accuracy of UAS-collected data with that obtained from TLS, which serves as a benchmark. The results reveal that the UAS demonstrates a relative accuracy Root Mean Square Error (RMSE) of approximately ±4.1 cm compared to TLS. Concerning absolute precision, the UAS-produced RMSE values are determined as ±0.20719 for the X coordinate, ±0.164 for the Y coordinate, and ±0.001584 for the Z coordinate compared to GNSS data, which functions as the benchmark. The utilization of UAS technology offers a non-invasive measurement approach that eliminates direct physical contact between the operator and the documented object. This non-intrusive method ensures the preservation of the original scene characteristics and has shown its superiority over conventional approaches in managing crash scenes. Overall, this study underscores the potential of UAS technology in accurately reconstructing crime scenes for forensic investigation purposes.

Open Set Vehicle Detection for UAV-Based Images Using an Out-of-Distribution Detector

Vehicle target detection is a key technology for reconnaissance unmanned aerial vehicles (UAVs). However, in order to obtain a larger reconnaissance field of view, this type of UAV generally flies at a higher altitude, resulting in a relatively small proportion of vehicle targets in its imaging images. Moreover, due to the unique nature of the mission, previously unseen vehicle types are prone to appearing in the surveillance area. Additionally, it is challenging for large-scale detectors based on deep learning to achieve real-time performance on UAV computing equipment. To address these problems, we propose a vehicle object detector specifically designed for UAVs in this paper. We have made modifications to the backbone of Faster R-CNN based on the target and scene characteristics. We have improved the positioning accuracy of small-scale imaging targets by adjusting the size and ratio of anchors. Furthermore, we have introduced a postprocessing method for out-of-distribution detection, enabling the designed detector to detect and distinguish untrained vehicle types. Additionally, to tackle the scarcity of reconnaissance images, we have constructed two datasets using modeling and image rendering techniques. We have evaluated our method on these constructed datasets. The proposed method achieves a 96% mean Average Precision at IoU threshold 0.5 (mAP50) on trained objects and a 71% mAP50 on untrained objects. Equivalent flight experiments demonstrate that our model, trained on synthetic data, can achieve satisfactory detection performance and computational efficiency in practical applications.

Exploit Domain-Robust Optical Flow in Domain Adaptive Video Semantic Segmentation

Domain adaptive semantic segmentation aims to exploit the pixel-level annotated samples on source domain to assist the segmentation of unlabeled samples on target domain. For such a task, the key is to construct reliable supervision signals on target domain. However, existing methods can only provide unreliable supervision signals constructed by segmentation model (SegNet) that are generally domain-sensitive. In this work, we try to find a domain-robust clue to construct more reliable supervision signals. Particularly, we experimentally observe the domain-robustness of optical flow in video tasks as it mainly represents the motion characteristics of scenes. However, optical flow cannot be directly used as supervision signals of semantic segmentation since both of them essentially represent different information. To tackle this issue, we first propose a novel Segmentation-to-Flow Module (SFM) that converts semantic segmentation maps to optical flows, named the segmentation-based flow (SF), and then propose a Segmentation-based Flow Consistency (SFC) method to impose consistency between SF and optical flow, which can implicitly supervise the training of segmentation model. The extensive experiments on two challenging benchmarks demonstrate the effectiveness of our method, and it outperforms previous state-of-the-art methods with considerable performance improvement. Our code is available at https://github.com/EdenHazardan/SFC.

An ETA-Based Tactical Conflict Resolution Method for Air Logistics Transportation

Air logistics transportation has become one of the most promising markets for the civil drone industry. However, the large flow, high density, and complex environmental characteristics of urban scenes make tactical conflict resolution very challenging. Existing conflict resolution methods are limited by insufficient collision avoidance success rates when considering non-cooperative targets and fail to take the temporal constraints of the pre-defined 4D trajectory into consideration. In this paper, a novel reinforcement learning-based tactical conflict resolution method for air logistics transportation is designed by reconstructing the state space following the risk sectors concept and through the use of a novel Estimated Time of Arrival (ETA)-based temporal reward setting. Our contributions allow a drone to integrate the temporal constraints of the 4D trajectory pre-defined in the strategic phase. As a consequence, the drone can successfully avoid non-cooperative targets while greatly reducing the occurrence of secondary conflicts, as demonstrated by the numerical simulation results.

Intelligent Computing Collaboration for the Security of the Fog Internet of Things

The application of fog Internet of Things (IoT) technology helps solve the problem of weak computing power faced by IoT terminals. Due to asymmetric differences in communication methods, sensing data offloading from IoT terminals to fog and cloud layers faces different security issues, and both processes should be protected through certain data transmission protection measures. To take advantage of the relative asymmetry between cloud, fog, and sensing layers, this paper considers using physical layer security technology and encryption technology to ensure the security of the sensing data unloading process. An efficient resource allocation method based on deep reinforcement learning is proposed to solve the problem of channel and power allocation in fog IoT scenarios, as well as the selection of unloading destinations. This problem, which is NP-hard, belongs to the attribute of mixed integer nonlinear programming. Meanwhile, the supporting parameters of the method, including state space, action space, and rewards, are all adaptively designed based on scene characteristics and optimization goals. The simulation and analysis show that the proposed method possesses good convergence characteristics. Compared to several heuristic methods, the proposed method reduces latency by at least 18.7% on the premise that the transmission of sensing data is securely protected.