Sliding Window Detection Research Articles

Visual Detection of Lost Ear Tags in Breeding Pigs in a Production Environment Using the Enhanced Cascade Mask R-CNN

As the unique identifier of individual breeding pigs, the loss of ear tags can result in the loss of breeding pigs’ identity information, leading to data gaps and confusion in production and genetic breeding records, which can have catastrophic consequences for breeding efforts. Detecting the loss of ear tags in breeding pigs can be challenging in production environments due to factors such as overlapping breeding pig clusters, imbalanced pig-to-tag ratios, and relatively small-sized ear tags. This study proposes an improved method for the detection of lost ear tags in breeding pigs based on Cascade Mask R-CNN. Firstly, the model utilizes ResNeXt combined with a feature pyramid network (FPN) as the feature extractor; secondly, the classification branch incorporates the online hard example mining (OHEM) technique to improve the utilization of ear tags and low-confidence samples; finally, the regression branch employs a decay factor of Soft-NMS to reduce the overlap of redundant bounding boxes. The experiment employs a sliding window detection method to evaluate the algorithm’s performance in detecting lost ear tags in breeding pigs in a production environment. The results show that the accuracy of the detection can reach 92.86%. This improvement effectively enhances the accuracy and real-time performance of lost ear tag detection, which is highly significant for the production and breeding of breeding pigs.

A Spatio-Temporal Spotting Network with Sliding Windows for Micro-Expression Detection

Micro-expressions reveal underlying emotions and are widely applied in political psychology, lie detection, law enforcement and medical care. Micro-expression spotting aims to detect the temporal locations of facial expressions from video sequences and is a crucial task in micro-expression recognition. In this study, the problem of micro-expression spotting is formulated as micro-expression classification per frame. We propose an effective spotting model with sliding windows called the spatio-temporal spotting network. The method involves a sliding window detection mechanism, combines the spatial features from the local key frames and the global temporal features and performs micro-expression spotting. The experiments are conducted on the CAS(ME)2 database and the SAMM Long Videos database, and the results demonstrate that the proposed method outperforms the state-of-the-art method by 30.58% for the CAS(ME)2 and 23.98% for the SAMM Long Videos according to overall F-scores.

Spatially-Coupled Faster-Than-Nyquist Signaling: A Joint Solution to Detection and Code Design

In this paper, we investigate two important issues of faster-than-Nyquist (FTN) signaling, namely, reduced-complexity detection and code design. Different from previous works, we consider these two issues jointly by designing a scheme that increases the minimum squared Euclidean distance of FTN signaling via repetition coding at a cost of an increased complexity. Furthermore, to reduce the rate loss of the repetition, we adopt the idea of spatially-coupling from coding theory to FTN signaling, and the resultant signaling scheme is therefore referred to as spatially-coupled faster-than-Nyquist (SC-FTN) signaling. The signal of SC-FTN signaling is generated in a continuous manner by interleaving and repeating the coded FTN signals and a graph-based iterative sliding-window detector is applied for signal detection. Both bounding and extrinsic information transfer chart analysis are provided to study the error-floor and convergence performances of SC-FTN signaling. These analyses unveil the intrinsic relationship between error floor, decoding threshold, and detection/decoding complexity, which provides guidelines for the designs of practical systems. Simulation results show that the promising error performance can be achieved with a simple FTN detection, where the bit error rate performance of coded SC-FTN signaling outperforms that of state-of-the-art coded FTN systems and the capacity of Nyquist signaling.

Sliding Window Detection and Analysis Method of Night-Time Light Remote Sensing Time Series—A Case Study of the Torch Festival in Yunnan Province, China

The spatial distribution of night-time lights (NTL) provides a new perspective for studying the range and influence of human activities. However, most studies employing NTL time series are based on monthly or annual composite data, and time series studies incorporating sliding windows are currently lacking. Therefore, using National Polar-Orbiting Partnership’s visible infrared imaging radiometer suite (NPP-VIIRS) night-time light remote sensing (NTLRS) data, VNP46A2, toponym, and Yunnan census statistical data, this study proposes a sliding-window-based NTLRS time series detection and analysis method. We extracted ethnic minority areas on the PyCharm platform using ethnic minority population proportion data and toponym and excluding data representing interference from urban areas. We used a sliding window approach to analyze NTLRS time series data of each ethnic group and calculated the cosine similarity between the NTL brightness curve of original data and the sliding window analysis result. The cosine similarity was greater than 0.96 from 2018 to 2020; we also conducted a field trip to the 2019 Torch Festival to demonstrate the applicability of the employed method. Finally, the temporal and spatial pattern of the Torch Festival was analyzed using the festival in Yunnan Province as an example. Results showed that the Torch Festival, mostly celebrated by the Yi ethnic group, was usually held on the 24th (and ranged from the 22nd to 26th) day in the sixth month of the lunar calendar (LC) every year. We found that during the Torch Festival, the greater the increase in the percentage of NTL brightness reduction in the main urban area of Kunming, the greater the percentage of ethnic minorities’ NTL brightness. The width of the sliding window can be adjusted appropriately according to the research objective, with these results showing good continuity. Our study presents a new application of the sliding window approach in the field of remote sensing, suitable for research into festivals related to night lights and fire all over the world.

A supervised approach for automated surface defect detection in ceramic tile quality control

Surface defect detection is very important to guarantee the quality of ceramic tiles production. At present, this process is usually performed manually in the ceramic tile industry, which is low efficiency and time-consuming. For small surface defects detection of high-resolution ceramic tiles image, an intelligent detection method for surface defects of ceramic tiles based on an improved you only look once version 5 (YOLOv5) algorithm is presented. Firstly, the high-resolution ceramic tile images are cropped into slices, and the Bottleneck module in the YOLOv5s network is optimized by introducing depthwise convolution and replaced in the whole network. Then, feature extraction is performed using the improved Shufflenetv2 backbone, and an attention mechanism is added to the backbone network to improve the feature extraction ability. The path aggregation network (PAN) and Feature Pyramid Networks (FPN) neck are used to enhance the feature extraction, and finally, the YOLO head is used to identify and locate the ceramic tile defects. The multiple sliding windows detection method is proposed to detect the original ceramic tile image which is faster than the single sliding window detection method. The experimental results show that compared with the original YOLOv5s detection algorithm, the parameters of the model are reduced by 20.46 %, the floating point operations are reduced by 26.22 %, and the mean average precision (mAP) of the proposed method is 96.73 % in the ceramic tile image slice test set which has 1.93 % improvement in mAP than the original YOLOv5s. Compare with other object detection methods, the method proposed in this paper also has certain advantages. In the high-resolution ceramic tile images test set, the mAP of the proposed algorithm is 86.44 % by using the multiple sliding window detection method. The ceramic defect detection experiment has verified the feasibility of the method proposed in this paper.

A Joint User Activity Detection and Channel Estimation Scheme for Packet-Asynchronous Grant-Free Access

Grant-free (GF) access has been viewed as a promising solution to meet the low-latency requirement in massive access of short packets due to the cancellation of the procedure of requesting for the grant. While most GF schemes are realized in a synchronous way, a recently proposed packet-asynchronous GF (PA-GF) scheme can further reduce the overall delay. However, its performance was only validated under the ideal conditions, i.e., perfect active user detection (AUD) and channel estimation (CE). In this letter, in order to realize timely AUD and CE for PA-GF practically, a novel frame structure for multiplexing pilot sequences and data is proposed together with a sliding window detection algorithm. A modification of the orthogonal matching pursuit (OMP)—the prior-aided OMP (PA-OMP)—is also proposed. Numerical results show that the proposed algorithm achieves the low-latency advantage of PA-GF under imperfect AUD and CE conditions.

Research on the Application of Virtual Technology-based Posture Detection Device in Swimming Teaching

With the socio-economic development, the national demand for playing leisure sports has increased, and swimming is one of the popular choices. To help swimming beginners understand the correct swimming posture more quickly and directly, hybrid neural network algorithms based on sliding window detection and deep residual networks are designed in this study, and two corresponding virtual image classification models of swimmer’s posture are designed based on these algorithms. In order to reduce the noise of the input data and reduce the cost of data collection, the virtual reality technology is used to convert the swimmer’s swimming pose image into the image model in the virtual reality space as the input data of the algorithm. The performance test experimental results show that the classification accuracy of the swimmer pose recognition models based on PTP-CNN algorithm and SW-CNN algorithm designed in this research are 97.48% and 96.72% respectively on the test set, which are much higher than other comparison models, and the model built based on PTP-CNN algorithm has the fastest computation speed. The results of this research can be applied to assist participants in swimming pose recognition in teaching beginner swimmers.

Obstacle Avoidance and Environmental Adaptability Analysis of Snake-like Robot Based on Deep Learning

Aiming at the problems of high complexity and low accuracy of visual depth map feature recognition, a graph recognition algorithm based on principal component direction depth gradient histogram (pca-hodg) is designed in this study. In order to obtain high-quality depth map, it is necessary to calculate the parallax of the visual image. At the same time, in order to obtain the quantized regional shape histogram, it is necessary to carry out edge detection and gradient calculation on the depth map, then reduce the dimension of the depth map combined with the principal component, and use the sliding window detection method to reduce the dimension again to realize the feature extraction of the depth map. The results show that compared with other algorithms, the pca-hodg algorithm designed in this study improves the average classification accuracy and significantly reduces the average running time. This shows that the algorithm can reduce the running time by reducing the dimension, extract the depth map features more accurately, and has good robustness.

Moving-Vehicle Identification Based on Hierarchical Detection Algorithm

The vehicle detection method plays an important role in the driver assistance system. Therefore, it is very important to improve the real-time performance of the detection algorithm. Nowadays, the most popular method is the scanning method based on sliding window search, which detects the vehicle from the image to be detected. However, the existing sliding window detection algorithm has many drawbacks, such as large calculation amount and poor real-time performance, and it is impossible to detect the target vehicle in real time during the motion process. Therefore, this paper proposes an improved hierarchical sliding window detection algorithm to detect moving vehicles in real time. By extracting the region of interest, the region of interest is layered, the maximum and minimum values of the detection window in each layer are set, the flashing frame generated by the layering is eliminated by the delay processing method, and a method suitable for the motion is obtained: the real-time detection algorithm of the vehicle, that is, the hierarchical sliding window detection algorithm. The experiments show that the more layers are divided, the more time is needed, and when the number of detection layers is greater than 7, the time change rate increases significantly. As the number of layers decreases, the detection accuracy rate also decreases, resulting in the phenomenon of a false positive. Therefore, it is determined to meet the requirements of real time and accuracy when the image is divided into 7 layers. It can be seen from the experiment that when the images to be detected are divided into 7 layers and the maximum and minimum values of detection windows are 30 × 30 and 250 × 250, respectively, the number of sub-windows generated is one thirty-seventh of the original sliding window detection algorithm, and the execution time is only one-third of the original sliding window detection algorithm. This shows that the hierarchical sliding window detection algorithm has better real-time performance than the original sliding window detection algorithm.

Bayesian estimate of system availability for consecutive k-out-of-n:F system

In the efficient design and functionality of complex systems, redundancy problems in systems play a key role. The consecutive-k-out-of-n:F structure, which has broad application in street light arrangements, vacuum systems in an accelerator, sliding window detection, relay stations for a communication system. Availability is one of the significant measures for a maintained device because availability accounts for the repair capability. A very significant feature is the steady-state availability of a repairable device. For the repairable consecutive k-out-of-n:F system with independent and identically distributed components, the Bayesian point estimate (B.P.E) of steady-state availability under squared error loss function (SELF) and confidence interval are obtained.

Two-Stage Degradation Assessment and Prediction Method for Aircraft Engine Based on Data Fusion

Aeroengine is one of the most concerned objects of the relevant aviation industry and researchers, and it is a hard work to assess and predict performance degradation due to the complex structure and the changeable operating condition of the engine. In order to realize the performance degradation assessment and remaining useful life (RUL) prediction of aeroengine, this paper proposes a two-stage assessment and prediction method based on data fusion. First, the standard deviation merged by multiple selected features is used as the health indicator to characterize the engine performance. Second, a sliding window detection method called average local window slope is proposed to determine the current health state of observations by a specified rule. Finally, the RUL prediction is performed on the observation in the two stages, respectively. On the one hand, a similarity-based RUL prediction method is used to engines in the health stage, and on the other hand, for engines in the degradation stage, a RUL prediction method based on a mapping function of the standard deviation and the current using cycle is established. The proposed method has been applied and verified on the NASA’s C-MAPSS simulation data. Results of degradation assessment and prediction show that the proposed method is trustworthy and feasible from the engineering perspective, and it has better performance in the comprehensive indicator compared with other methods.

Accurate and Lightweight RailNet for Real-Time Rail Line Detection

Railway transportation has always occupied an important position in daily life and social progress. In recent years, computer vision has made promising breakthroughs in intelligent transportation, providing new ideas for detecting rail lines. Yet the majority of rail line detection algorithms use traditional image processing to extract features, and their detection accuracy and instantaneity remain to be improved. This paper goes beyond the aforementioned limitations and proposes a rail line detection algorithm based on deep learning. First, an accurate and lightweight RailNet is designed, which takes full advantage of the powerful advanced semantic information extraction capabilities of deep convolutional neural networks to obtain high-level features of rail lines. The Segmentation Soul (SS) module is creatively added to the RailNet structure, which improves segmentation performance without any additional inference time. The Depth Wise Convolution (DWconv) is introduced in the RailNet to reduce the number of network parameters and eventually ensure real-time detection. Afterward, according to the binary segmentation maps of RailNet output, we propose the rail line fitting algorithm based on sliding window detection and apply the inverse perspective transformation. Thus the polynomial functions and curvature of the rail lines are calculated, and rail lines are identified in the original images. Furthermore, we collect a real-world rail lines dataset, named RAWRail. The proposed algorithm has been fully validated on the RAWRail dataset, running at 74 FPS, and the accuracy reaches 98.6%, which is superior to the current rail line detection algorithms and shows powerful potential in real applications.

Voice activity detection using optimal window overlapping especially over health-care infrastructure

PurposeThis paper aims to introduce recently an unregulated unsupervised algorithm focused on voice activity detection by data clustering maximum margin, i.e. support vector machine. The algorithm for clustering K-mean used to solve speech behaviour detection issues was later applied, the application, therefore, did not permit the identification of voice detection. This is critical in demands for speech recognition.Design/methodology/approachHere, the authors find a voice activity detection detector based on a report provided by a K-mean algorithm that permits sliding window detection of voice and noise. However, first, it needs an initial detection pause. The machine initialized by the algorithm will work on health-care infrastructure and provides a platform for health-care professionals to detect the clear voice of patients.FindingsTimely usage discussion on many histories of NOISEX-92 var reveals the average non-speech and the average signal-to-noise ratios hit concentrations which are higher than modern voice activity detection.Originality/valueResearch work is original.

Node-Subset Scheduling with a Subset-Reset Mechanism for the Decoding Algorithm of Nonbinary LDPC Codes

For the conventional extended min-sum (EMS) algorithm, all check nodes update their check-to-variable (C2V) messages in every iteration. Selected scheduling, which reduces the number of check nodes for message updating in one iteration, can effectively reduce the complexity of the decoding algorithm, but it also lead to some performance degradation. With the introduction of a metric based on node stability, we propose stability-based node-subset scheduling (SNS) for the EMS algorithm, which can improve the performance of node-subset scheduling (NS). Second, to further improve the decoding performance of SNS while maintaining low complexity, we propose the SNS-EMS algorithm with a subset-reset mechanism (RSNS-EMS) based on the abnormal stability found in the processing node subset, which will cause the estimated codeword to fail to converge. The RSNS-EMS algorithm enhances performance through a sliding window detection and reset mechanism, and it resets the elements in the processing node subset to force all check nodes to update new messages when abnormal stability is detected. The simulation results show that the proposed algorithm can reduce complexity by approximately 25% with negligible performance degradation.

Radar CFAR detection for multiple-targets situations for Weibull and log-normal distributed clutter

In the presence of Weibull clutter, the development of sliding window detection processes, based on scale and power invariant distributions, has been extensively examined. This involves the selection of two functions labeled as scale-invariant and secondary CRP (clutter range profile) functions. However, due to the presence of outliers, existing CFAR (constant false alarm rate) algorithms show remarkable CFAR losses. We resort in this work to the practice of a suitable choice of these two functions in order to have a new decision rule with immunity against interfering targets. To do this, a quadruple-order statistics-based CFAR detection algorithm with different four-order statistics are proposed in the presence of log-normal and Weibull clutter disturbances. Via Monte Carlo simulations, the analysis of the false alarm regulation of the proposed detector is studied showing its robustness with respect to clutter parameters. Moreover, for comparison purposes with existing CFAR algorithms, simulated results indicate that lowest CFAR losses can be obtained by the proposed quadruple-order statistics named WHWH-CFAR (Weber–Haykin-Weber–Haykin) in the presence of strong interfering targets. IPIX real data are also performed to test the validity of the proposed detector.

Safety Cruise Control of Connected Vehicles Using Radar and Vehicle-to-Vehicle Communication

It is a common traffic phenomenon that side vehicles change lanes without any forewarning, which is easy to induce vehicle side collision and rear-end collision. In order to improve the active safety of cruise control in such a random traffic environment, this article proposes a connected cruise control (CCC) framework considering merging behavior probability of side vehicles. The control strategy of the framework first uses vehicle-to-vehicle communication and radar recognition algorithm to detect the surrounding vehicle targets. Then, the merging potential of the side vehicle is determined through the combination of fuzzy support vector machine and sliding window detection. Afterward, a cruise distance strategy considering the cutting-in probability of merging behavior is proposed and applied to the CCC system to realize the early braking of the main vehicle. Finally, the longitudinal control of the host vehicle is realized by the acceleration control, considering both the time delay and feedback error. The effectiveness of the proposed systematic control framework is verified through not only simulation but also field test. Results show that our method can accurately detect the lane change intent of the side vehicle, thus realizing braking of the host vehicle in advance when the side vehicle changes lane.

Multi-layered tree crown extraction from LiDAR data using graph-based segmentation

With the development of Light Detection and Ranging (LiDAR) and Unmanned Aerial Vehicle (UAV) technology, extracting tree crowns from LiDAR data and infering their geometrical features are becoming more available for everyone. Although the forest has a significant stratification phenomenon in the vertical direction, the existing individual tree detection methods generally aimed at solving overstory trees detection. As a result, the understory trees cannot be effectively extracted. To effectively detect individual tree from multi-layer forests, a multi-layered tree extraction method using a graph-based segmentation algorithm was proposed in this study. First, using the graph-based segmentation algorithm delineates the canopy of the overstory tree on the canopy height model (CHM) generated by LiDAR data. Then, using the sliding window detection method extracts the LiDAR data of understory trees. Finally, the information of understory trees is extracted by the graph-based segmentation algorithm. To verify the performance of the proposed method, this study selected six experimental plots from two research areas. According to the result of our method, the highest matching score and average score for overstory trees reached 91.3 and 86.3; the highest matching score and average score for understory trees reached 78.1 and 63.2. Compared with other methods, our method has better detection results. The experimental results show that the proposed method can extract the understory and overstory trees effectively, thus improving the accuracy of individual tree extraction in multi-layered forests.

A Deep Learning-Aided Detection Method for FTN-Based NOMA

The rapid booming of future smart city applications and Internet of things (IoT) has raised higher demands on the next-generation radio access technologies with respect to connection density, spectral efficiency (SE), transmission accuracy, and detection latency. Recently, faster-than-Nyquist (FTN) and nonorthogonal multiple access (NOMA) have been regarded as promising technologies to achieve higher SE and massive connections, respectively. In this paper, we aim to exploit the joint benefits of FTN and NOMA by superimposing multiple FTN-based transmission signals on the same physical recourses. Considering the complicated intra- and interuser interferences introduced by the proposed transmission scheme, the conventional detection methods suffer from high computational complexity. To this end, we develop a novel sliding-window detection method by incorporating the state-of-the-art deep learning (DL) technology. The data-driven offline training is first applied to derive a near-optimal receiver for FTN-based NOMA, which is deployed online to achieve high detection accuracy as well as low latency. Monte Carlo simulation results validate that the proposed detector achieves higher detection accuracy than minimum mean squared error-frequency domain equalization (MMSE-FDE) and can even approach the performance of the maximum likelihood-based receiver with greatly reduced computational complexity, which is suitable for IoT applications in smart city with low latency and high reliability requirements.

Interference control in sliding window detection processes using a Bayesian approach

Recently a Bayesian approach has been introduced, for the design of non-coherent sliding window detection processes, to operate in X-band high resolution maritime surveillance radar scenarios. The classical problem of comparing a cell under test with a normalised measurement of the clutter level is re-interpreted in terms of a Bayesian predictive inference approach. Viewing unknown clutter parameters as statistical variables, it is possible to construct a Bayesian posterior distribution, using Jeffreys priors. This approach has permitted the construction of detection processes, with the constant false alarm rate property, in situations where this was previously difficult to achieve. However, in the presence of interference in the clutter cells, the detector degrades in performance, as do others constructed in more traditional ways. Here it will be shown how the Bayesian approach can be extended to produce sliding window detectors which are immune to the presence of interference.

Detecting regional dominant movement patterns in trajectory data with a convolutional neural network

ABSTRACT Detecting movement patterns with complicated spatial or temporal characteristics is a challenge. The past decade has witnessed the success of deep learning in processing image, voice and text data. However, its application in movement pattern detection is not fully exploited. To address the research gap, this paper develops a deep learning approach to detect regional dominant movement patterns (RDMP) in trajectory data. Specifically, a novel feature descriptor called directional flow image (DFI) is firstly proposed to store the local directional movement information in trajectory data. A DFI classification model called TRNet is designed based on convolutional neural network. The model is then trained with a synthetic trajectory dataset covering 11 classes of commonly encountered movement patterns in reality. Finally, a sliding window detector is built to detect RDMP at multiple scales and a clustering-based merging method is proposed to prune the redundant detection results. Training of TRNet on the synthetic dataset achieves considerably high accuracy. Experiments on a real-world taxi trajectory dataset further demonstrate the effectiveness and efficiency of the proposed approach in discovering complex movement patterns in trajectory data.