Real-time Performance Research Articles

Adaptive occlusion object detection algorithm based on OL-IoU

The continuous advancement of autonomous driving technology imposes higher demands on the accuracy of target detection in complex environments, particularly when traffic targets are occluded. Existing algorithms still face significant challenges in detection accuracy and real-time performance under such conditions. To address this issue, this paper proposes an improved YOLOX algorithm based on adaptive deformable convolution, named OCC-YOLOX. This algorithm enhances the feature extraction network’s ability to focus on occluded targets by incorporating a coordinate attention mechanism. Additionally, it introduces the Overlapping IoU (OL-IoU) loss function to optimize the overlap between predicted and ground truth bounding boxes, thereby improving detection accuracy. Furthermore, the adoption of Fast Spatial Pyramid Pooling (Fast SPP) reduces computational complexity while maintaining real-time performance. Experiments on fused public datasets demonstrate that OCC-YOLOX achieves improvements in accuracy, recall, and average precision by 2.76%, 1.25%, and 1.92%, respectively. In addition to testing on the KITTI, CityPersons, and BDD100K datasets, the effectiveness of the OCC-YOLOX algorithm is further validated through comparisons with self-collected occlusion scene data. The experimental results indicate that OCC-YOLOX outperforms existing mainstream detection algorithms, particularly in handling complex occlusion scenarios, significantly enhancing the accuracy and efficiency of object detection. This study provides new insights for addressing the challenges of occluded target detection in intelligent transportation systems.

Comprehensive Analysis of Cloud Computing Performance Factors: Investigating the Impact of Response Time, Load Balancing and Service Broker Policies on Cloud Service Efficiency Using CloudSim Simulation

Cloud performance refers to the efficiency and effectiveness with which a cloud system operates delivering hosted services over the internet. As cloud computing continues to offer flexibility, scalability and computational power monitoring and improving cloud performance is essential. Performance optimization is influenced by factors such as load balancing and service broker policies which impact system response times and overall user experience. This paper provides an in-depth review of key publications and real-time cloud performance tools identifying critical performance factors that affect cloud efficiency. Notably, response time emerged as a fundamental metric for cloud service quality. Using CloudSim simulation we examine cloud performance evaluation criteria and experimentally assess the impact of response time dependencies on broker policies, load balancing techniques and data center distribution. This study offers a framework for understanding cloud performance evaluation and highlights strategies to enhance user experience in diverse cloud environments.

A Novel Embedded Side Information Transmission Scheme Based on Polar Code for Peak-to-Average Power Ratio Reduction in Underwater Acoustic OFDM Communication System.

In this paper, we proposed an embedded side information (SI) transmission scheme based on polar code construction for PAPR reduction using the PTS scheme in the underwater Acoustic (UWA) Orthogonal Frequency Division Multiplexing (OFDM) communication system. We use polar codes due to the ability of the arbitrarily designed code rate. Additionally, polar codes can be employed to establish a nested code structure consisting of multiple subsets. The SI bits can be embedded in a polar codeword by exploiting these features. Thus, the approach does not occupy existing data rates or cause additional loss in data transmission rates. At the same time, it embeds m-sequence into the polar code as an indicator vector for the blind SI detector, which makes the blind SI detector able to autonomously discriminate SI at the receiver. Simulation and tank experiment results indicate that the proposed embedded SI transmission scheme has the potential to significantly decrease the likelihood of whole-symbol error caused by SI errors. Meanwhile, the proposed PTS scheme eliminates the need to wait for the entire packet to be received before obtaining the SI, thereby preventing waste of data storage devices and ensuring real-time performance of the Underwater Acoustic Communication (UAC) OFDM system. This achieves symbol-level real-time calculation for the system.

RS-DETR: An Improved Remote Sensing Object Detection Model Based on RT-DETR

Object detection is a fundamental task in computer vision. Recently, deep-learning-based object detection has made significant progress. However, due to large variations in target scale, the predominance of small targets, and complex backgrounds in remote sensing imagery, remote sensing object detection still faces challenges, including low detection accuracy, poor real-time performance, high missed detection rates, and high false detection rates in practical applications. To enhance remote sensing target detection performance, this study proposes a new model, the remote sensing detection transformer (RS-DETR). First, we incorporate cascaded group attention (CGA) into the attention-driven feature interaction module. By capturing features at different levels, it enhances the interaction between features through cascading and improves computational efficiency. Additionally, we propose an enhanced bidirectional feature pyramid network (EBiFPN) to facilitate multi-scale feature fusion. By integrating features across multiple scales, it improves object detection accuracy and robustness. Finally, we propose a novel bounding box regression loss function, Focaler-GIoU, which makes the model focus more on difficult samples, improving detection performance for small and overlapping targets. Experimental results on the satellite imagery multi-vehicles dataset (SIMD) and the high-resolution remote sensing object detection (TGRS-HRRSD) dataset show that the improved algorithm achieved mean average precision (mAP) of 78.2% and 91.6% at an intersection over union threshold of 0.5, respectively, which is an improvement of 2.0% and 1.5% over the baseline model. This result demonstrates the effectiveness and robustness of our proposed method for remote sensing image object detection.

Visual and Radar Data Fusion: Complementary or Alternative Sensors in Intelligent Driving Systems

This article explores the role and advantages and disadvantages of visual and radar data fusion in intelligent driving systems. With the rapid development of intelligent driving technology, the comprehensive perception of the environment by car perception systems has become crucial, usually requiring the combination of low, medium, and high-precision sensors. Visual sensors perform excellently in acquiring high-resolution images and recognizing detailed information, while radar sensors have higher stability and penetration in harsh weather conditions. Due to the limitations of a single sensor in complex driving environments, data fusion has become a key strategy for improving the performance of perception systems. By comparing different perception schemes and application scenarios, this article analyzes the actual effects and potential advantages of visual and radar data fusion, revealing its important role in intelligent driving technology. Fusion technology can compensate for the shortcomings of various sensors, and improve the accuracy and reliability of environmental perception, but still faces challenges in data synchronization, algorithm selection, and real-time performance. This article aims to provide valuable insights for the development of future intelligent driving technology.

Modeling and Transition Flight Control of Distributed Propulsion–Wing VTOL UAV with Induced Wing Configuration

The integration of propulsion and wing in distributed propulsion–wing UAVs (DPW UAVs) introduces significant propulsion-aerodynamic coupling, complicating dynamic modeling and flight control. This complexity is heightened by using induced wing surfaces for vertical takeoff and landing, requiring controllers to adapt to configuration changes and disturbances during transition flight. This paper develops a propulsion-aerodynamic coupling model for a medium-sized DPW UAV with induced wings (DPW-IW), enabling real-time aerodynamic performance calculations. Furthermore, a unified flight-control framework is proposed to avoid controller scheduling and switching during flight mode transitions. The proposed control framework employs the time-scale separation principle, divided into an outer loop and an inner loop. The outer loop uses a fuzzy controller to adjust allocation parameters, while the inner loop applies incremental nonlinear dynamic inversion (INDI) and control allocation (INCA) methods, providing robustness to nonlinear changes during flight transitions. Finally, simulations under various conditions demonstrate the controller’s effectiveness in ensuring smooth and robust transitions.

Lightweight defect detection algorithm of tunnel lining based on knowledge distillation

Due to the influence of construction quality, engineering geology and hydrological environment, defects such as dehollowing and insufficient compaction can occur in tunnels. Aiming at the problems of complex detection model, poor real-time performance and low accuracy of the current tunnel lining defect detection methods, the study proposes a lightweight defect detection algorithm of tunnel lining based on knowledge distillation. Firstly, a high-precision teacher model based on yolov5s was constructed by constructing a C3CSFM module that combines residual structure and attention mechanism, a MDFPN network structure with multi-scale feature fusion and a reweighted RWNMS re-screening mechanism. Secondly, in the distillation process, the feature and output dimension results are fused to improve the detection accuracy, and the mask feature relationship is learned in the space and channel dimension to improve the real-time detection. Tests on the tunnel lining radar defect image dataset showed that the number of parameters of the improved model was reduced from 16.03 MB to 3.20 MB, a reduction of 80%, and the average accuracy was improved from 83.4 to 86.5%, an increase of 3.1%. On the basis of maintaining the structure and detection performance of the model, the lightweight degree of the model is greatly improved, and the high-precision and real-time detection of tunnel lining defects is realized.

BronchoTrack: Airway Lumen Tracking for Branch-Level Bronchoscopic Localization.

Localizing the bronchoscope in real time is essential for ensuring intervention quality. However, most existing vision-based methods struggle to balance between speed and generalization. To address these challenges, we present BronchoTrack, an innovative real-time framework for accurate branch-level localization, encompassing lumen detection, tracking, and airway association. To achieve real-time performance, we employ benchmark light weight detector for efficient lumen detection. We firstly introduce multi-object tracking to bronchoscopic localization, mitigating temporal confusion in lumen identification caused by rapid bronchoscope movement and complex airway structures. To ensure generalization across patient cases, we propose a training-free detection-airway association method based on a semantic airway graph that encodes the hierarchy of bronchial tree structures. Experiments on 11 patient datasets demonstrate BronchoTrack's localization accuracy of 81.72%, while accessing up to the 6th generation of airways. Furthermore, we tested BronchoTrack in an in-vivo animal study using a porcine model, where it localized the bronchoscope into the 8th generation airway successfully. Experimental evaluation underscores BronchoTrack's real-time performance in both satisfying accuracy and generalization, demonstrating its potential for clinical applications.

Event-Driven Maximum Correntropy Filter Based on Cauchy Kernel for Spatial Orientation Using Gyros/Star Sensor Integration.

Gyros/star sensor integration provides a potential method to obtain high-accuracy spatial orientation for turntable structures. However, it is subjected to the problem of accuracy loss when the measurement noises become non-Gaussian due to the complex spatial environment. This paper presents an event-driven maximum correntropy filter based on Cauchy kernel to handle the above problem. In this method, a direct installation mode of gyros/star sensor integration is established and the associated mathematical model is derived to improve the turntable's control stability. Based on this, a Cauchy kernel-based maximum correntropy filter is developed to curb the influence of non-Gaussian measurement noise for enhancing the gyros/star sensor integration's robustness. Subsequently, an event-driven mechanism is constructed based on the filter's innovation information for further reducing the unnecessary computational cost to optimize the real-time performance. The effectiveness of the proposed method has been validated by simulations of the gyros/star sensor integration for spatial orientation. This shows that the proposed filtering methodology not only has strong robustness to deal with the influence of non-Gaussian measurement noise but can also achieve superior real-time spatial applications with a small computational cost, leading to enhanced performance for the turntable's spatial orientation using gyros/star sensor integration.

Vibration displacement measurement of bridge structural models using image super-resolution reconstruction and visual object detection network

Abstract Visual vibration measurement has emerged in the field of structural health monitoring in recent years, but it still has some shortcomings in terms of resolution, recognition rate and real-time performance. Considering the three aspects of recovering high-frequency image details, improving the compactness of the target bounding box, and reducing the computational time, we use the constructed image super-resolution reconstruction model and target detection model to measure the vibration displacement of the bridge structural model. First, we integrate the Transformer module into the Unet network with a simple structure. The Swin and Global Transformer Unet (SGTU) module constructed in this form can reduce the computational cost while reconstructing the large-resolution feature map target, and it can sharply edge information of the vibration target. We use the framework of the YOLOv5 algorithm as the backbone, and use the GhostBottleneck (GB) module to reduce the time for convolution operations to generate similar features. In addition, the proposed DWCBottleneck (DWCB) fusion module is also able to achieve high-level semantic fusion and network depth expansion with minimal computational cost. Finally, the center point offset of the bounding box predicted by the model can be used to obtain the displacement offset of the object in the image sequence. The position information of the target in the first frame image is used as the reference frame for calculating the offset, and the vibration displacement of the flexible structure in the image coordinate system is obtained by calculating the deviation of the displacement between the remaining frames and the first frame. We perform qualitative and quantitative comparisons in three aspects: video super-resolution reconstruction, visual detection robustness, and sensor vibration measurement displacement using a homemade vibration image dataset. The time–frequency domain displacement curves regressed by the visual vibration measurement algorithm are compared with the curves acquired after accelerometer acquisition, indicating the necessity of super-resolution reconstruction in visual vibration measurement.

Santa María Academy: Shaping the Heroes of Today with Tomorrow's Technology

Santa María Academy has successfully integrated advanced technology into its educational framework to enhance the learning experience of its students, preparing them for future careers in law enforcement and the military. The academy's innovative approach includes personalized coaching, interactive resources, and smart tools that cater to individual student needs. By leveraging artificial intelligence and real-time performance monitoring, Santa María Academy has significantly improved student outcomes, motivation, and engagement. This case study explores the key elements of their success, including the implementation of personalized learning plans, interactive podcasts, a smart syllabus, and a real-time performance dashboard. The academy's model serves as a benchmark for educational institutions aiming to merge traditional values with cutting-edge technology to produce well-rounded and highly skilled graduates.

An adaptive online monitoring system using NFSOGI-PLL for three-phase voltage unbalance in grids

The extensive utilization of unbalanced and nonlinear loads has significantly impacted power quality in modern grids. Among various metrics, the three-phase voltage unbalance factor (TPVUF) stands out as a crucial indicator. To mitigate the challenges posed by harmonics and DC components that interfere with accurate TPVUF measurements, this paper introduces the Notch Filter-based Second-Order Generalized Integrator Phase-Locked Loop (NFSOGI-PLL) for the first time and designs an adaptive online monitoring system for TPVUF based on NFSOGI-PLL. To validate its performance, the hardware design and rapid software development of the NFSOGI-PLL-based TPVUF online monitoring system are thoroughly implemented, and the system is tested under various conditions. The results indicate that the proposed system not only provides real-time performance and high accuracy, complying with the rigorous standards set by GB/T 15543-2008 and IEC 61000-4-27, but also demonstrates remarkable stability under diverse influences. These findings underscore the exceptional filtering and tracking abilities of the NFSOGI-PLL in power systems, as well as the accuracy and anti-interference capabilities of the NFSOGI-based TPVUF online monitoring system, which holds great potential for widespread application in power system control and research.

Digital Twin Used in Real-Time Monitoring of Operations Performed on CNC Technological Equipment

This article presents the development and implementation of a real-time monitoring solution designed for CNC machines, specifically applied to 150 industrial printing machines, leveraging Digital Twin (DT) technology. The system integrates an SQL database with Android and .NET interfaces, ensuring seamless data synchronization across all machines and optimizing production processes. The real-time monitoring enables immediate reflection of operational changes, enhancing predictive maintenance and reducing machine downtime. A notable feature of the system is its 1 s average data synchronization rate per machine, managing 150 resources distributed over a 10,000 mp area. This fast synchronization improves workflow coordination, reducing production time by approximately 10%, and minimizing operator delays caused by material issues, machine malfunctions, or product defects. The integration of advanced analytics further supports real-time decision-making, predictive maintenance, and performance optimization, aligning the solution with the objectives of Industry 4.0 and Industry 5.0 initiatives. This version reflects the specific results of the research, including the 1 s synchronization rate, the 10% reduction in production time, and the scalability of the system for 150 resources.

High-Precision Measurement Method for Small Angles Based on the Defect Spot Mode of the Position-Sensitive Detector.

The paper proposes and verifies a small-angle measurement method based on the defect spot mode of the position-sensitive detector (PSD). With the output characteristics of the PSD in the defect spot mode and the size transformation properties of a focused beam, the measurement sensitivity can be significantly improved. Calibration experiments with the piezoelectric transducer (PZT) indicate that compared with the current PSD-based autocollimation method, the proposed method can improve the sensitivity of small-angle measurement by 57 times, and the measurement sensitivity of the proposed method can be further improved by optimizing the system parameters, while the proposed method has the advantages of a simple system and high real-time performance. Therefore, the proposed method is expected to be used in high-precision motion error detection, as well as in shape and position measurement.

Research on improving the ranging accuracy of ships with stereo vision through Kalman filter optimization.

The real-time and accurate detection and ranging of ships play a pivotal role in ensuring navigation safety, this study aims to enhance the navigation safety and environmental perception capabilities of inland waterway vessels. In the ship detection stage, addressing challenges such as large parameters, high computational complexity, and poor real-time performance in existing ship detection models, this paper proposes the MS-YOLOv5s ship target detection algorithm. This algorithm, based on YOLOv5s, utilizes the lightweight MobileNetV3-Small network to replace the original YOLOv5s feature extraction backbone network, thereby improving the detection speed. The results indicate that the parameter size of the MS-YOLOv5s model is 3.55M, only 50.49% of YOLOv5s. Achieving a detection rate of 50.28 FPS, the precision is 96.80%, and the mAP is 98.40%, striking a balance between high accuracy and low computational demand. In the depth estimation stage, influenced by the environment, leading to unstable measurement data, this paper proposes a binocular Kalman filter fusion ranging algorithm. The standard deviation of the ranging results is minimized to 6.032μm, which is one order of magnitude smaller than traditional ranging algorithms, significantly enhancing the robustness of the measurement results. Within a distance of 20m from the ship target, the error can be controlled within 3%, showcasing the applicability of the method proposed in this paper in complex inland waterway environments contributes to the enhancement of ships' environmental perception capabilities and navigation safety, holding positive implications for the development of intelligent vessels.

RTPL: A Real-Time Communication Protocol for LoRa Network

The industrial Internet of Things (IIoT) is prominently emerging in applications of large-scale and wide-area applications, such as oilfield management, smart grid management, real-time equipment monitoring, and integration of traffic management systems for smart cities. Relying on short-range wireless technologies (e.g., WirelessHART and ISA100.11a), traditional wireless solutions for industrial automation find it challenging to support the expansive scale of today’s IIoT. To address this limitation, we propose to adopt LoRaWAN, a prominent low-power wide-area network technology, for industrial automation. LoRaWAN for industrial automation poses some unique challenges. The fundamental building blocks of any industrial automation system are feedback control loops that largely rely on real-time communication. LoRaWAN traditionally adopts a simple protocol based on ALOHA with no collision avoidance or Listen Before Talk with Clear Channel Assessment and Random Backoff mechanisms to minimize energy consumption, which are less suitable for real-time communication. Existing real-time protocols for short-range technologies cannot be applied to a LoRaWAN network due to its unique characteristics such as asymmetry between downlink and the uplink spectrum, predefined modes (or classes) of operation, and concurrent reception through orthogonal spreading factors. In this paper, we address these challenges and propose RTPL - a Real-Time communication Protocol for LoRaWAN networks. RTPL is a low-overhead and conflict-free communication protocol allowing autonomous real-time communication of low-energy devices and exploits LoRa’s capability of parallel communication. We implement our approach on LoRa devices and evaluate through both physical experiments and extensive simulations. All results show that RTPL achieves on average 75% improvement in real-time performance without sacrificing throughput or energy compared to traditional LoRaWAN.

Real-Time Leaves Segmentation in RGB Images with Deep Learning in a Single-Board Computer

Abstract. This work proposed and evaluated methods for real-time leaf segmentation using a single-board computer. The main aim was to explore the state-of-the-art techniques based on the YOLO algorithm for real-time operation. For this purpose, the available variants of YOLOv8 and YOLOv9 were evaluated, and a semi-automatic labelling method based on the Segment Anything Model (SAM) algorithm was used. Given the need to delimit the leaf contour for labelling, it was possible to create a larger and more accurate dataset compared to the purely manual procedure. In addition, the cost-benefit of the applied algorithms and methods were assessed, considering the computational demand required, as well as the accuracy, recall, and precision delivered by these techniques. In this study, both quantitative analysis of the trained architectures' metrics and qualitative examination through direct observation of images were conducted to identify crucial aspects. The experiments were conducted with a post-processed dataset and the suitability for real-time applications was based on the elapsed time for segmentation. We concluded that the YOLOv8n architecture is the best one among those tested, presenting a precision and recall of 0.9064 and 0.7233, respectively. This architecture represents the best cost-benefit ratio between computational cost and real-time performance, being able to perform segmentation in 310 ms with the NVIDIA Jetson Nano board. Furthermore, when computational cost is not a problem or even when segmentation time can be higher, the YOLO8m network may be recommended when the recall metric is more important than precision. This network presented a precision and recall of 0.8556 and 0.7726, respectively, and presented a better performance in segmenting leaves located in more complex parts of the image and with a higher recall.

Crop Classification from Drone Imagery Based on Lightweight Semantic Segmentation Methods

Technological advances have dramatically improved precision agriculture, and accurate crop classification is a key aspect of precision agriculture (PA). The flexibility and real-time nature of UAVs have led them to become an important tool for acquiring agricultural data and enabling precise crop classification. Currently, crop identification relies heavily on complex high-precision models that often struggle to provide real-time performance. Research on lightweight models specifically for crop classification is also limited. In this paper, we propose a crop classification method based on UAV visible-light images based on PP-LiteSeg, a lightweight model proposed by Baidu. To improve the accuracy, a pyramid pooling module is designed in this paper, which integrates adaptive mean pooling and CSPC (Convolutional Spatial Pyramid Pooling) techniques to handle high-resolution features. In addition, a sparse self-attention mechanism is employed to help the model pay more attention to locally important semantic regions in the image. The combination of adaptive average pooling and the sparse self-attention mechanism can better handle different levels of contextual information. To train the model, a new dataset based on UAV visible-light images including nine categories such as rice, soybean, red bean, wheat, corn, poplar, etc., with a time span of two years was created for accurate crop classification. The experimental results show that the improved model outperforms other models in terms of accuracy and prediction performance, with a MIoU (mean intersection ratio joint) of 94.79%, which is 2.79% better than the original model. Based on the UAV RGB images demonstrated in this paper, the improved model achieves a better balance between real-time performance and accuracy. In conclusion, the method effectively utilizes UAV RGB data and lightweight deep semantic segmentation models to provide valuable insights for crop classification and UAV field monitoring.

An intelligent monitoring approach for urban natural gas pipeline leak using semi-supervised learning generative adversarial networks

Traditional gas pipeline leak monitoring methods are subjected to the long response times and high false alarm rates. Deep learning can enhance the accuracy and real-time performance of pipeline leak monitoring. This paper develops an intelligent monitoring approach for urban gas pipeline leaks based on a semi-supervised learning Generative Adversarial Network (SGAN). First, the Isolation Forest algorithm is used to classify anomalies in the collected process parameter data of urban natural gas pipelines. One-Hot Encoding is used to label a small amount of sample data of pipeline leak. Second, both the labeled and unlabeled data are input into SGAN model for semi-supervised learning and classification to monitor the state of urban gas pipeline leak. The methodology addresses the imbalance between pipeline leak status data and normal data. The comparison with GAN and MLP shows that the methodology reaches the highest values in all evaluation metrics (precision = 94.1%, accuracy = 95.63%, recall = 93.93%, F1 score = 94.32%). The superior performance and accuracy make it more effective for urban natural gas pipeline leak monitoring.

Application of 5G Remote Robotic-Assisted Laparoscopy in Urological Surgery: A Small Sample Analysis

ObjectiveTo evaluate the practical application and clinical safety of 5G remote robotics in urological surgery. MethodsWe conducted a comprehensive analysis of 14 cases of 5G remote domestic robot-assisted laparoscopic urological surgeries performed at our center from May 2023 to June 2024. Clinical data and network information were collected, and metrics such as operative time, blood loss, and complication rate were analyzed. We assessed the stability, accuracy, and safety of remote operations, as well as the acceptance and satisfaction of both doctors and patients with this technology. ResultsThe 14 surgeries conducted at our center involved a master-slave distance of 52 kilometers, with an average operative time of 83.3 minutes, an average blood loss of 23 milliliters, a conversion-to-open-surgery rate of 0%, and no complications. During the surgeries, the 5G remote robotic system functioned stably, with a clear field of vision, an average download speed of 216.5 Mbps, an average upload speed of 86.6 Mbps, an average maximum latency of 129.3 ms, an average minimum latency of 20.7 ms, and a packet loss rate of 0%. These results indicate that the system can provide a highly accurate operational view and effective transfer of surgical skills. Both doctors and patients expressed general satisfaction with this technology, recognizing its significant contribution to the safety and effectiveness of the surgery. ConclusionThe 5G remote robot-assisted laparoscopic urological surgeries conducted at our center demonstrate that combining 5G technology with robotic ensures the stability, real-time performance, and safety of remote surgeries.