检索结果-内蒙古大学图书馆

A three-stage detection algorithm for automatic crack-width identification of fine concrete cracks

JOURNAL OF CIVIL STRUCTURAL HEALTH MONITORING 2024年第6期14卷 1373-1382页

作者： Huang, Huang Wu, Zhishen Shen, Haifeng Ibaraki Univ Dept Urban & Civil Engn Hitachi 3168511 Japan Southeast Univ Int Inst Urban Syst Engn Nanjing Peoples R China

Semantic image segmentation is extensively used for automatic concrete crack detection. In previous studies on semantic image segmentation, concrete images were usually labeled as crack and noncrack zones, and recognition models were then trained using artificial neural networks. However, there is not enough edge information in concrete images for the trained model to identify effectively fine concrete cracks (widths < 0.1 mm). Furthermore, complex backgrounds in concrete images can cause false detections. To improve efficiency and reduce false detections, this study develops a three-stage automatic crack-width identification method for fine concrete cracks. First, a full crack skeleton information identification based on image segmentation is proposed. The performance of the mainstream image segmentation architectures, PSP-Net, Seg-Net, U-Net, and Res-Unet, are compared and analyzed, demonstrating that the Res-Unet-based crack skeleton segmentation is the most accurate at fine-crack detection and able to solve the information loss problem that occurs when learning the imbalanced data of fine concrete cracks. Second, a fractal dimension (FD)-based false detection removal process is applied to discriminate true cracks and false detections. The results show that false detections (line-like curves, shadows, and surface stains) can be removed, increasing the matching rate from 0.6476 to 0.8351. Finally, the FD features of the crack skeleton with maximum widths < 0.1 mm, crack widths in the range of 0.1-0.2 mm, and crack widths > 0.2 mm are calculated. Findings illustrate that the values of the FD feature for the three crack-width ranges are suitable for quantitative characterization of identified crack widths.

关键词： detection algorithm Image segmentation Concrete crack-width identification Artificial neural network

来源：评论

学校读者我要写书评

暂无评论

Novel rogue optical network unit detection algorithm for gigabit passive optical networks

引用

OPTICA APPLICATA 2016年第3期46卷 421-433页

作者： Horvath, Tomas Munster, Petr Dubravec, Lubos Filka, Miloslav Brno Univ Technol Fac Elect Engn & Commun Dept Telecommun Tech 12 Brno 61600 Czech Republic Orange Slovensko As Metodova 8 Bratislava 82108 Slovakia

Passive optical networks are widely used as a promising solution for future access networks. Currently, the bandwidth is still increasing which means the current copper networks are not able to transfer new services such as 4K video, live streaming, etc. In other words, they reached their capacity limit. The passive optical networks rely on point-to-multipoint technology. That means each customer uses a share medium by time slots. Each time slot exactly specifies who and when is able to transfer data. In general, this control mechanism is implemented in the optical network unit by worst transmission convergence layer. On the other hand, there are cases when the optical network unit (it is called rogue optical network unit) does not follow instructions provided by the optical line termination, for example, if an attacker modifies a firm ware of the end unit and/or when the control protocol is not loaded properly inside optical network unit. In worst case, the optical network unit transmits data in a continual mode (other optical network units cannot send data). The standard defines finding of the rogue optical network unit but it does not specify how the rogue optical network unit should be allocated because the frames of the rogue optical network unit do not contain the proper parameters. We realized a measurement in a real network with the rogue optical network unit and then we analyzed the captured data. A new algorithm for the rogue optical network unit allocation is presented. We do not consider any modification of the transmission convergence layer in gigabit passive optical networks.

关键词： gigabit passive optical network (GPON) rogue optical network unit (ONU) data analysis detection algorithm

来源：评论

学校读者我要写书评

暂无评论

Extraction of Trap Energy Distribution in Proton-Irradiated Focal Plane Arrays Using a High-Precision Random Telegraph Noise detection algorithm

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2021年 70卷 1-10页

作者： Pepel, Victor D. Webster, Preston T. Logan, Julie, V Morath, Christian P. Appl Technol Associates Albuquerque NM 87123 USA Air Force Res Lab Space Vehicles Directorate Albuquerque NM 87117 USA MIT Dept Nucl Sci & Engn Cambridge MA 02139 USA

Random telegraph noise (RTN) is a type of low-frequency noise that degrades the sensitivity of focal plane arrays (FPAs) and other electronics. Careful analysis of RTN provides information about the material defects generating this noise and ultimately may lead to a solution to eliminate this noise source from the optoelectronic imaging system. The focus of this effort is the definition and demonstration of a simple histogram-based detection algorithm, which evaluates the shape of the pixel output histogram in order to detect RTN pixels and extract the energy levels of the defects. The algorithm sensitivity and precision statistics are benchmarked as a function of the detection threshold. Applying the algorithm to a 1280 x 10(24) infrared FPA irradiated with 3.2 x 10(11) protons/cm(2) (63 MeV) shows a maximum 84% increase in the number of pixels exhibiting RTN from preradiation to post-300 K anneal. The distribution of preradiation trap energies shows clusters at 13 meV above the Fermi level and 14 and 44 meV below the Fermi level. The radiation-induced defects, however, exhibit a different energy distribution with an increased concentration at the Fermi level, suggesting different energy distributions for the traps created during the manufacturing process compared to those created by proton radiation.

关键词： Defects detection algorithm focal plane array (FPA) proton radiation random telegraph noise (RTN) trap energy extraction

来源：评论

学校读者我要写书评

暂无评论

Convolutional-Neural-Network-Based detection algorithm for Uplink Multiuser Massive MIMO Systems

引用

IEEE ACCESS 2020年 8卷 64250-64265页

作者： Li, Lin Hou, Huijun Meng, Weixiao Nanjing Univ Posts & Telecommun Jiangsu Key Lab Broadband Wireless Commun & Inter Nanjing 210046 Peoples R China Nanjing Univ Posts & Telecommun Sch Internet Things Nanjing 210046 Peoples R China CETC 14th Res Inst Nanjing 210039 Peoples R China Harbin Inst Technol Commun Res Ctr Harbin 150001 Peoples R China

In the coming 6th generation (6G) and beyond in wireless communication, an increasing number of ultrascale intelligent factors, including mobile robot users and smart cars, will result in interference exploitation. The management of this exploitation will be a great challenge for detection algorithms in uplink massive multiple-input and multiple-output (MIMO) systems, especially for high-order quadrature amplitude modulation (QAM) signals. Artificial intelligence technology employing machine learning is one of the key approaches among the 6G technical solutions. In this paper, a convolutional-neural-network-based likelihood ascent search (CNNLAS) detection algorithm is proposed on the basis of a graphical detection model for uplink multiuser massive MIMO systems. Compared with other algorithms, the proposed CNNLAS detection algorithm has a stronger robustness against the channel estimation errors, and requires lower average received signal-to-noise ratios to obtain better bit error rate performance and to achieve the theoretical spectral efficiency with a lower polynomial average per symbol computational complexity, both for the graphical low-order and high-order QAM signals in uplink multiuser massive MIMO systems.

关键词： detection algorithms Uplink Massive MIMO Transmitting antennas Quadrature amplitude modulation Bit error rate Wireless communication massive multiple-input and multiple-output (MIMO) convolutional neural network (CNN) detection algorithm high-order modulation bit error rate (BER) computational complexity

来源：评论

学校读者我要写书评

暂无评论

Crop-row detection algorithm based on Random Hough Transformation

引用

MATHEMATICAL AND COMPUTER MODELLING 2011年第3-4期54卷 1016-1020页

作者： Ji, Ronghua Qi, Lijun China Agr Univ Coll Informat & Elect Engn Beijing 100083 Peoples R China

It is important to detect crop rows accurately for field navigation. In order to spray on line, a variable rate spray system should detect the crop center line accurately. Most existing detection algorithms are slow to detect crop rows because of the complicated calculation. The gradient-based Random Hough Transform algorithm could improve the calculation speed and reduce the computation effectively by the more-to-one merger mapping method. In order to detect the center of the crop row rapidly and effectively, the detection algorithm with gradient-based Random Hough Transform was proposed to detect the center line of crop rows. We tested the center line of crop-row detection for three kinds of plant distribution, being sparse, general and intensive. The experimental results showed that the detection algorithm with gradient-based Random Hough Transform was adaptive to the difference of plant density in the crop row effectively. Contrasted with the detection algorithm based on the Hough transform, the detection algorithm based on the gradient-based Random Hough was faster and had a high detection correction rate. (C) 2010 Elsevier Ltd. All rights reserved.

关键词： detection algorithm Random Hough Transform Crop row

来源：评论

学校读者我要写书评

暂无评论

Optimized detection algorithm of complex intrusion interference signal in mobile wireless network

引用

JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY 2018年第3期21卷 771-779页

作者： Zhang, Yi Chengdu Med Coll Sch Humanities & Informat Management Chengdu 610500 Sichuan Peoples R China

The current interference signal detection algorithm can not effectively suppress noise, and reduce the detection ability of intrusion interference signal. In this paper, an optimized detection algorithm for complex intrusion interference signals in mobile wireless network is proposed. Signal analysis algorithm is used for intrusion signal detection in mobile wireless network. Then, according to the noise suppression process, the single frequency signal is discretized and analyzed, and the multi-component non-stationary signal model of network intrusion is constructed. Finally, the empirical mode algorithm is used to separate the harmonic components of the intrusion interference signal in mobile wireless network, so as to realize the optimal detection of the intrusion interference signal in mobile wireless network. The simulation results show that the proposed algorithm can effectively suppress the noise generated by the signal, and is sensitive to the detection ability of the signal, and has good detection ability.

关键词： Mobile wireless network Intrusion interference signal detection algorithm

来源：评论

学校读者我要写书评

暂无评论

Analysis of Data Point Cloud Preprocessing and Feature Angle detection algorithm

引用

RECENT ADVANCES IN ELECTRICAL & ELECTRONIC ENGINEERING 2021年第7期14卷 700-707页

作者： Zhao, Feng Dhiman, Gaurav Wuhan Business Univ Modern Educ Technol Ctr Wuhan 430056 Hubei Peoples R China Govt Bikram Coll Commerce Dept Comp Sci Patiala Punjab India

Background: The two main stages are utilized for feature extraction, from which the first stage consists of a penalty weight to the neighbor graph's edges. The edge penalty weights are minimized by the neighbor sub-graph extraction to produce the set of feature patterns. For noisy data, the second stage is helpful. Methodology: In order to realize the measurement of the geometric dimensions of the ship block, this paper uses the theory of computer vision and reverse engineering to obtain the data of the segmented-hull with the method of digitizing the physical parts based on the vision, and processes the data by using the relevant knowledge of reverse engineering. Result: The results show that the efficiency of the edge extraction algorithm based on mathematical morphology is 30% higher than that of the mesh generation method. An adaptive corner detection algorithm based on the edge can adaptively determine the size of the support area and accurately detect the corner position. Conclusion: According to the characteristics of the point cloud of ship hull segment data, an adaptive corner detection algorithm based on the edge is adopted to verify its feasibility.

关键词： Data preprocessing corner detection reverse engineering segmented hull detection algorithm edge extraction

来源：评论

学校读者我要写书评

暂无评论

Mature Tomato Fruit detection algorithm Based on improved HSV and Watershed algorithm

引用

IFAC-PapersOnLine 2018年第17期51卷 431-436页

作者： Malik, Muhammad Hammad Zhang, Ting Li, Han Zhang, Man Shabbir, Sana Saeed, Ahmed Key Laboratory of Modern Precision Agriculture System Integration Research Ministry of Education China Agricultural University Haidian Beijing China Beijing Beida Software Engineering Co. Ltd. China College of Information and Electrical Engineering China Agricultural University HaidianBeijing China

Tomato is well known fruit since it has many essential and beneficial nutrients like antioxidant, vitamin C and A for human daily diet. Tomato picking by hand is both labor and time consuming. Therefore, to overcome these issues, tomato needs to be picked up automatically with the help of harvesting robot. Recently automation of fruit harvesting gains great popularity. To guide the harvesting robot to pick up the fruit correctly, it is important to correctly detect and find the location of the red mature fruit. In this study, a new mature tomato detection algorithm based on the improved HSV (Hue, Saturation, Value) color space and the improved watershed segmentation was proposed, by using a cheap and easily accessible RGB camera for guiding a robot to pick up red tomatoes automatically. For that purpose, the color images of tomato plants were captured under natural light. At the first stage, improved HSV transform was used to remove background and to detect only red tomatoes. Then the morphological operations were applied to modify the detected fruit. The second step was the separation of connected fruits, so that the harvesting robot can pick up fruit correctly. To achieve this task, the improved watershed algorithm was applied. As a result, the detection of separated red mature tomato was done by using the improved HSV and watershed, with uneven illumination and complex background. The overall red fruit detection accuracy was up to 81.6%, and it shows good potential to be used on harvesting robot. © 2018

关键词： Fruits Color Computer vision Error detection Harvesting Mathematical morphology Pickups Robots Signal detection Watersheds detection accuracy detection algorithm Harvesting robot Localization Morphological operations Uneven illuminations Water shed algorithm Watershed segmentation

来源：评论

学校读者我要写书评

暂无评论

YOLOX-Dense-CT: a detection algorithm for cherry tomatoes based on YOLOX and DenseNet

引用

JOURNAL OF FOOD MEASUREMENT AND CHARACTERIZATION 2022年第6期16卷 4788-4799页

作者： Zheng, Hao Wang, Guohui Li, Xuchen Xian Technol Univ Sch Optoelect Engn Xian 710021 Peoples R China

The detection of cherry tomatoes has a great significance for robotic harvesting. However, uneven environment conditions, such as branch and leaf occlusion, cluster of fruits, and so on, have made the cherry tomatoes detection very challenging. This paper proposes an effective cherry tomato detection algorithm called YOLOX-Dense-CT to solve these problems. To be specific, the DenseNet network is treated as the basic backbone of the original YOLOX to make the whole network suitable for the cherry tomatoes. Moreover, the convolutional block attention module (CBAM) attention mechanism is applied to make the features from the backbone more fused with the Neck part. As suggested by the experimental results, the proposed YOLOX-Dense-CT model is effective in detecting cherry tomatoes, with the mean average precision (mAP) reaching 94.80%, which is 4.02% higher than the original YOLOX-L model. Meanwhile, the number of parameters is only 34.6 M, which is 19.6 M lower than the YOLOX-L. Furthermore, the YOLOX-Dense-CT is compared to general target detection models and it has the best detection performance. In summary, the proposed method can well meet the requirements of high accuracy detection and provide a strategy for the cherry tomato detection system.

关键词： Cherry tomatoes Robotic harvesting detection algorithm YOLOX DenseNet

来源：评论

学校读者我要写书评

暂无评论

Clinical experience with a new detection algorithm for differentiation of supraventricular from ventricular tachycardia in a dual-chamber defibrillator

引用

JOURNAL OF CARDIOVASCULAR ELECTROPHYSIOLOGY 2004年第6期15卷 646-652页

作者： Sinha, AM Stellbrink, C Schuchert, A Möx, B Jordaens, L Lamaison, D Gill, J Kaplan, A Merkely, B Univ Klinikum Aachen Med Klin 1 D-52057 Aachen Germany Univ Klin Hamburg Eppendorf Hamburg Germany Univ Rotterdam Hosp Rotterdam Netherlands Univ Klin Greifswald Greifswald Germany HRU Hop Gabriel Montpied Clermont Ferrand France St Thomas Hosp London England Tri City Cardiol Mesa AZ USA Semmelweis Univ H-1085 Budapest Hungary

Introduction: Inadequate therapy for supraventricular tachyarrhythmias (SVT) is a frequent problem of implantable cardioverter defibrillators (ICD). Dual-chamber ICDs have been developed to improve discrimination of SVT from ventricular tachycardia (VT). We investigated the positive predictivity, sensitivity, and specificity of a new algorithm, the SMART detection(TM) algorithm, incorporated in the Phylax AV (Biotronik) dual-chamber ICD. Methods and Results: Two hundred nine patients (185 men, age 64 11 years) received a Phylax AV ICD with SMART detection(TM) activated. In 138 of these patients, 1,245 sustained tachycardia episodes with a detailed electrogram were stored in the device during a follow-up period of 10 6 months. Episodes were correctly classified as ventricular fibrillation (VF, n = 178) in 52 patients, VT (n = 641) in 98 patients, and SVT (n = 385) in 48 patients by the algorithm. Forty-one true SVT episodes (3.3%) were misclassified as VT: atrial fibrillation (n = 7) and flutter (n = 1), sinus tachycardia (n = 12), and other SVT (n = 21). The positive predictivity for VF/VT was 94.5% (95% CI 92.7-95.8) uncorrected and 94.5 % (95% CI 92.995.8%) corrected with the generalized equation estimation (GEE) method. The positive predictivity for SVT was 100%. The specificity was 88.9% (95% CI 85.6-91.6%) uncorrected and 89.0% (95% CI 85.6-91.6%) corrected with the GEE method with a sensitivity of 100%. Conclusion: The SMART detection (TM) algorithm was safe and reliable for the detection of all ventricular tachycardias. Although its specificity was high, it should be improved with regard to SVT to avoid inappropriate ICD therapies.

关键词： implantable cardioverter defibrillator dual chamber arrhythmia detection algorithm inappropriate therapy optimization of programming

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：