Research on an anti-crawling mechanism and key algorithm based on sliding time window

Vehicle detection is an important component in unmanned driving systems. This paper presents a real-time vehicle detection method under complex road conditions to solve the problem of real-time detection of road vehicles in automatic driving. Firstly using yolo network to build a deeper depth neural network, which is used to identify and score the vehicles in the image. Then, using the dynamic threshold method to remove some false candidate boxes, and using the Gauss attenuation function to filter the overlapping candidate boxes. The loss function normalized by the change rate is used to train the neural network, and the batch normalization layer is used to correct the input data of the network to avoid data deviation during the training process. Finally, a full convolution layer is added at the end of the network layer to transform the two-dimensional data into one-dimensional data and output the final recognition results. It is verified that this method improves the efficiency and accuracy of real-time vehicle detection, it can effectively detect vehicles on roads with complex backgrounds, and satisfy the real-time requirements of road vehicle detection.

Tiny object detection is an important and challenging object detection subfield. However, many of its numerous applications (e.g., human tracking and marine rescue) have tight detection time constraints. Namely, two-stage object detectors are too slow to fulfill the real-time detection needs, whereas one-stage object detectors have an insufficient detection accuracy. Consequently, enhancing the detection accuracy of one-stage object detectors has become an essential aspect of real-time tiny objects detection. This work presents a novel model for real-time tiny objects detection based on a one-stage object detector YOLOv5. The proposed YOLO-P4 model contains a module for detecting tiny objects and a new output prediction branch. Next, a weighted bi-directional feature pyramid network (BiFPN) is introduced in YOLO-P4, yielding an improved model named YOLO-BiP4 that enhances the YOLO-P4 feature input branches. The proposed models were tested on the Tiny-Person dataset, demonstrating that the YOLO-BiP4 model outperforms the original model in detecting tiny objects. The model satisfies the real-time detection needs while obtaining the highest accuracy compared to existing one-stage object detectors.

Deep Learning is the latest research achievement in the field of Artificial Intelligence. To detect and identify the state of circuit breaker in ambient air automatic monitor station, Deep Learning was proposed to realize the circuit breaker state recognition in real-time. The Deep Convolutional Neural Network was used to construct the state detection and identification model of the circuit breaker and provided the detection process. The remote monitoring system of ambient air automatic monitor station and inspection data were used to build the experimental data platform, with the circuit breaker state recognition data was acquired from the multiple ambient air automatic monitor stations. Through training the model with the Deep Learning, the experimental results showed that the circuit breaker state detection and identification method have a simple recognition process and high accuracy. The real-time Detection and Identification function of the circuit breaker state was realized, meanwhile, the assistant decision-making function was provided to the ambient air automatic monitor station.

Currently, the existing fault detection methods have problems in that the detection results are not in line with reality and the detection accuracy is too low when detecting the grounding circulation fault of transmission cables. This paper introduces the wavelet energy spectrum of module components to design and study the real-time fault detection method of grounding the circulating current of the transmission cable. In this study, the wavelet-based detection function that meets the requirements of fault detection is selected. On this basis, the traveling wave fault features are extracted by using the wavelet energy spectrum of the module component. Finally, the fault detection signal is decomposed in real-time, and the fault detection is realized according to the practical value of the reconstructed signal. The comparative experiment proves that the new detection method can get the detection result of the grounding current fault of the cable, which is more in line with the actual situation and has very high detection accuracy.

With the rapid development of Internet of Things (IoT) and UAV technology, for the whole IoT system of vehicle detection, the middle and high level of information transmission and server processing has made a breakthrough, so at the bottom of the real-time detection of the vehicle by the UAV is the key to the whole system. However, UAV vehicle detection faces the challenges of too many small targets in the image leading to low detection accuracy, limited hardware platform resources requiring control of model size, and real-time detection requiring high inference speed. Aiming at the above problems, we propose a lightweight model PrFu-YOLO based on YOLOv8 improvement, which achieves a good balance between the accuracy, inference speed, and model size. And it realizes real-time vehicle detection embedded in an UAV platform. To solve the problem of low vehicle detection accuracy, we design a new structure PrFuFPN based on adding a small target detection layer to achieve more advanced feature fusion. To address the limited resources of the platform and the problem of real-time vehicle detection, we add GhostConv to the structure and constantly try to adjust the parameters of the network. Finally, extensive experiments were conducted on the VisDrone2019 and CARPK data sets to fully evaluate the model. Compared to YOLOv8s on the VisDrone2019 test set, mAP50 was improved by 10.05%, mA95 by 14.54%, the number of parameters was reduced by 8.13%, and the model size was reduced by 12.5%, while the FPS of 67.13 fully met the needs of real-time detection.

Dynamic target recognition is an important issue in the field of image processing research. It is widely used in photoelectric detection, target tracking, video surveillance areas. Complex cruise scene of target detection, compared to the static background, since the target and background objects together and both are in motion, greatly increases the complexity of moving target detection and recognition. Based on the practical engineering applications, combining an embedded systems and real-time image detection technology, this paper proposes a real-time movement detection method on an embedded system based on the FPGA + DSP system architecture on an embedded system. The DSP digital image processing system takes high speed digital signal processor DSP TMS320C6416T as the main computing components. And we take large capacity FPGA as coprocessor. It is designed and developed a high-performance image processing card. The FPGA is responsible for the data receiving and dispatching, DSP is responsible for data processing. The FPGA collects image data and controls SDRAM according to the digital image sequence. The SDRAM realizes multiport image buffer. DSP reads real-time image through SDRAM and performs scene motion detection algorithm. Then we implement the data reception and data processing parallelization. This system designs and realizes complex cruise scene motion detection for engineering application. The image edge information has the anti-light change and the strong anti-interference ability. First of all, the adjacent frame and current frame image are processed by convolution operation, extract the edge images. Then we compute correlation strength and the value of movement offset. We can complete scene motion parameters estimation by the result, in order to achieve real-time accurate motion detection. We use images in resolution of 768 576 and 25Hz frame rate to do the real-time cruise experiment. The results show that the proposed system achieves real-time processing requirements for the engineering applications. 2014 SPIE.

Anomaly detection of flight data is vital to aviation safety. A real time anomaly detection for flight data can identify safety issues as soon as anomalies occur and improve the safety of aviation system. However, existing anomaly detection methods for flight data either perform not very well on unknown issues or are limited in their capabilities of real time detection. This paper provides a method to detect real time flight data anomalies with comparable detection performance. Firstly it transforms discrete and continuous flight data into discrete sequences. And each sequence obtained from discrete and continuous data is segmented into several subsequences by sliding a window of the specified size. More importantly, the variable length n-grams are obtained from each subsequence. Then all of unique n-grams become the feature space of the training data set. In this way each subsequence is mapped to this feature space and becomes an input feature vector of one-class Support Vector Machine (SVM). Finally, anomalies in flight data can be detected in real time by the classifier learned on these feature vectors corresponding to subsequences. Experimental results show that it performs better to use linear kernel to train one-class SVM on our feature vectors than Gaussian kernel and the size of the sliding window takes a little effect on the performance of our detection method. Moreover, this method outperforms Multiple Kernel Anomaly Detection (MKAD) in real time anomaly detection.

Environment perception is essential for autonomous driving technology. The curb is a prominent feature of urban roads and therefore is a significant part of environment perception. In this paper, a real-time curb detection and tracking method is proposed for Unmanned Ground Vehicles (UGVs). The proposed curb detection algorithm uses the the surrounding environment data provided by a 3D-LIDAR sensor to extract the curb position based on its spatial features. The curb tracking algorithm is proposed to predict and update the curb position with respect to the current vehicle states in real time. The performance of the proposed method is verified through extensive experiments with a UGV driving on campus roads. The experimental results demonstrate the accuracy and robustness of the proposed method.

The combination of UAV camera and intelligent algorithm is a promising method for non-contact bridge crack detection. In this paper, an inspection tool based on UAV Image Acquisition Technology (UAVIAT) and Improved Intelligent Target Detection Technology (IITDT) called Improved Intelligent Real-Time Crack Detection Method for Bridges (IIRTCDMB) is proposed for efficient crack detection. The contributions of this paper are (1) The Squeeze-Excitement (SE) attention module is integrated into the target detection algorithm - You Only Look Once version 7 (YOLOv7) model to improve the learning ability of the feature channel. A Focal-efficient intersection over union (Focal-EIoU) loss function is also introduced to improve the regression accuracy of the model. As a result, a new crack image detection algorithm, YOLOv7-CD, is proposed. (2) A training process based on two-stage transfer learning (TSTL) is established, and hyper-parameter optimization of YOLOv7-CD is carried out. The feasibility and excellent performance of the proposed method are verified by applying it on the Cuntan Yangtze River Bridge. The results show that the average precision (AP) of the YOLOv7-CD model is improved by 3.19% compared with the original YOLOv7 model. After TSTL and hyperparameter optimization, the AP of the YOLOv7-CD model for bridge crack detection reaches 98.01%, which is higher than that of the popular target detection models. The IIRTCDMB proposed in this paper can acquire bridge surface images more safely and efficiently, and provide inspectors with more accurate structural crack information with lower computational and hardware requirements, which can provide technical support for the assessment of structural safety conditions and the formulation of maintenance programs.

As one of the most common structural forms in port engineering, the operation environment of high-pile wharf is quite harsh and complex, and its pile foundation often produces structural damage of different degrees. Until now, there is a lack of efficient, safe and economic damage detection methods. A novel and precise real-time structural damage detection (SDD) method using both finite element modelling (FEM) and 1D convolutional neural networks (CNNs) is established in this study. The results indicate that the proposed method could accurately identify the presence and location of damage in real time. The results also demonstrated that the proposed 1D CNNs based model are more sensitive to the longitudinal and lateral displacement responses of the high-pile wharf structure.

Health monitoring and fault diagnosis of liquid rocket engine (LRE) are the most important concerning issue for the safety of rocket’s flying, especially for the man-carried aerospace engineering. Based on the sensor measurement signals of a certain type of hydrogen-oxygen rocket engine, this paper proposed a real-time fault detection approach using a genetic algorithm-based least squares support vector regression (GA-LSSVR) algorithm for the real-time fault detection of the rocket engine. In order to obtain effective training samples, the data is normalized in this paper. Then, the GA-LSSVR algorithm is derived through comprehensive considerations of the advantages of the Support Vector Regression (SVR) algorithm and Least Square Support Vector Regression (LSSVR). What is more, this paper provided the genetic algorithm to search for the optimal LSSVR parameters. In the end, the computational results of the suggested approach using the rocket practical experimental data are given out. Through the analysis of the results, the effectiveness and the detection accuracy of this presented real-time fault detection method using LSSVR GA-optimized is verified. The experiment results show that this method can effectively diagnose this hydrogen-oxygen rocket engine in real-time, and the method has engineering application value.

Walnut is one of the important economic tree species in the world. For the past few years, with the continuous expansion of walnut planting area, walnut bacterial blight caused by Xanthomonas campestrispv. juglandis has critically affected walnut production and caused economic losses. In this research, by comparing the gene sequences of X. campestrispv. juglandis and its closely related bacterial species, the 16S rDNA gene sequence was selected, and the specific primers were evaluated using PCR. Subsequently, the real-time fluorescence PCR specific detection method of X. campestrispv. juglandis was established and optimized. The specificity of real-time PCR was verified for 23 strains isolated from walnut and its surrounding soil, including eight target pathogens. The specific amplification was possible even when the cDNA template concentration was 7.0×10-5 µg/mL, hence, the detection sensitivity of the standard method is 7.0×10-5 µg/mL. The reliability of the established real-time PCR detection method in field detection was verified by the identification of target bacteria in 8 randomly collected healthy leaves and leaves of suspected walnut blight. This method laid a foundation for the early diagnosis of X. campestrispv. juglandis.

An improved tiny-yolov3 pedestrian detection algorithm

In this paper, we consider water surface object detection in natural scenes. Generally, background subtraction and image segmentation are the classical object detection methods. The former is highly susceptible to variable scenes, so its accuracy will be greatly reduced when detecting water surface objects due to the changing of the sunlight and waves. The latter is more sensitive to the selection of object features, which will lead to poor generalization as a result, so it cannot be applied widely. Consequently, methods based on deep learning have recently been proposed. The River Chief System has been implemented in China recently, and one of the important requirements is to detect and deal with the water surface floats in a timely fashion. In response to this case, we propose a real-time water surface object detection method in this paper which is based on the Faster R-CNN. The proposed network model includes two modules and integrates low-level features with high-level features to improve detection accuracy. Moreover, we propose to set the different scales and aspect ratios of anchors by analyzing the distribution of object scales in our dataset, so our method has good robustness and high detection accuracy for multi-scale objects in complex natural scenes. We utilized the proposed method to detect the floats on the water surface via a three-day video surveillance stream of the North Canal in Beijing, and validated its performance. The experiments show that the mean average precision (MAP) of the proposed method was 83.7%, and the detection speed was 13 frames per second. Therefore, our method can be applied in complex natural scenes and mostly meets the requirements of accuracy and speed of water surface object detection online.

To enhance the real-time performance and accuracy of damage detection in conveyor belts with poor image quality, insufficient sample feature information, and numerous scattered minor damages, we propose an improved visual detection method based on YOLOv5, incorporating Improved Depth-Separable Convolution (IDSC) and Scale Sequence Feature Fusion (SSFF). The IDSC method significantly reduces network parameters, achieving a lightweight model while maintaining detection accuracy, whereas the SSFF method effectively integrates spatial and multi-scale feature information, enhancing the identification of minor damage. Compared to existing methods, our approach provides a superior balance between detection speed and accuracy. The proposed IDSC-SSFF model achieves a mean average precision (mAP) of 95.65% while maintaining an inference speed of 139.1 FPS, meeting the stringent requirements of real-time defect detection in industrial environments.