检索结果-内蒙古大学图书馆

23rd IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

作者： Agarwal, Akshay Ratha, Nalini Noore, Afzel Singh, Richa Vatsa, Mayank IISER Bhopal Bhopal India SUNY Buffalo Buffalo NY USA Texas A&M Univ Kingsville TX USA IIT Jodhpur Jodhpur Rajasthan India

ISBN: (纸本)9781665493468

One of the critical steps in biometrics pipeline is detection of presentation attacks, a physical adversary. Several presentation (adversary) attack detection (PAD) algorithms, including iris PAD, have been proposed and have shown superlative performance. However, a recent study, on a small-scale database, has highlighted that iris PAD may have gender biases. In this research, we present a rigorous study on gender bias in iris presentation attack detection algorithms using a large-scale and gender-balanced database. The paper provides several interesting observations which can help in building future presentation attack detection algorithms with aim of fair treatment of each demography. In addition, we also present a robust iris presentation attack detection algorithm by combining gender-covariate based classifiers. The proposed robust classifier not only reduces the difference in accuracy between different genders but also improves the overall performance of the PAD system.

关键词： Computer vision Databases Demography Pipelines Buildings Classification algorithms detection algorithms

来源：评论

学校读者我要写书评

暂无评论

Automatic High-Frequency Click detection Algorithm for Cetaceans

Automatic High-Frequency Click Detection Algorithm for Cetac...

引用

International Symposium on Underwater Technology

作者： Meng-Fan Tsai Shu-Chun Chuang Shih-Hsien Weng Yin-Ying Fang Wei-Chun Hu Chi-Fang Chen Department of Engineering Science and Ocean Engineering National Taiwan University Taipei Taiwan R.O.C. Engineering department Oceansound CO. LTD. Taipei Taiwan R.O.C. Ship and Ocean Technology Research Center National Taiwan University Taipei Taiwan R.O.C.

ISBN: (数字)9798331534189

ISBN: (纸本)9798331534196

The expansion of offshore wind power facilities in Taiwan's western waters poses significant threats to marine cetaceans, particularly the finless porpoise (Neophocaena phocaenoides sunameri), which is the most frequently stranded cetacean in the region. Causes of stranding include fishery bycatch, environmental changes, vessel collisions, and marine engineering activities. Traditional visual surveys are ineffective for monitoring this species due to its lack of a prominent dorsal fin, limiting the collection of ecological data and the development of conservation strategies. To overcome these challenges, we developed an automated detection algorithm for high-frequency (HF) cetacean clicks using underwater passive acoustic monitoring (PAM). The first algorithm utilized a time-domain computational approach, applying a high-pass filter (>110 kHz) and a peak detection algorithm to identify click trains, incorporating the Peak Energy Algorithm (Step A). The second algorithm introduced a band-pass filter (60–192 kHz) and combined Step A with a Signal-to-Noise Ratio (SNR) calculation (Step B). Additionally, two filters were integrated into the algorithm, informed by studies conducted in the Bohai Sea, Zhangzhou, and Hainan, China, to refine click event detection. Analysis of 7.5 hours of acoustic data from the Yun-Chang Rise validated the algorithm. The first algorithm achieved a precision ratio (PR) of 78.09% and a false positive ratio (FPR) of 8.06%, with misclassifications attributed to broadband pulses or closely spaced clicks. The second algorithm, incorporating the SNR threshold, improved PR to 96.3% (+17.4%) and reduced FPR to 0.89% (−7.17%). This algorithm enables efficient HF cetacean click detection and can be integrated into firmware for real-time marine operations. It provides a promising solution to the limitations of visual surveys, enhancing acoustic pattern recognition, and supporting improved cetacean monitoring and conservation in Taiwan's waters.

关键词： Surveys Band-pass filters Visualization Acoustics Real-time systems Pattern recognition detection algorithms Monitoring Signal to noise ratio Microprogramming

来源：评论

学校读者我要写书评

暂无评论

Research on the detection Algorithm of Automobile Low Beam Feature Points Based on Point Symmetry Degree

Research on the Detection Algorithm of Automobile Low Beam F...

引用

Power Electronics, Computer Applications (ICPECA), IEEE International Conference on

作者： Shen Qiqi School of Cyberspace Security Changzhou College of Information Technology Changzhou China

ISBN: (数字)9798331533694

ISBN: (纸本)9798331533700

Accurate detection of the feature points of automobile low beam is a crucial step in the alignment and adjustment of automobile low beam headlights. Its detection accuracy plays a decisive role in the accuracy and reliability of the comprehensive detection results of automotive headlamps. Moreover, it is an important technical support for ensuring the safety and compliance of vehicle lighting systems. However, the existing detection algorithms for low beam feature points cannot generally and effectively detect and distinguish the feature points of low beam with different low beam light patterns. Therefore, this paper proposes a new detection algorithm for the feature points of automotive low beam light patterns based on point symmetry degree. The core of this algorithm lies in making full use of the significant difference in point symmetry degree between corner points and non-corner points, and realizing the precise detection of low beam feature points based on this feature. The experimental results show that this algorithm can accurately identify and distinguish the elbow points and other feature points in the currently common “two-segment” and “five-segment” low beam light patterns. It has high detection accuracy, demonstrates strong universality and robustness, can effectively resist various interference factors, and has broad application prospects in the fields of automobile headlight optical detection and alignment adjustment.

关键词： Accuracy Interference Resists Feature extraction Robustness Safety Automobiles detection algorithms Elbow Automotive engineering

来源：评论

学校读者我要写书评

暂无评论

Small target detection based on multimodal data fusion

Small target detection based on multimodal data fusion

引用

Computer Applications and Information Technology (ISCAIT), International Symposium on

作者： Haocheng Song College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China

ISBN: (数字)9798331542856

ISBN: (纸本)9798331542863

Object detection is a fundamental task in remote sensing with wide applications in autonomous vehicle navigation, surveillance, and search-and-rescue operations. Remote sensing images, typically captured by satellites, aircraft, and UAV sensors, are of great significance in both military and civilian domains due to their extensive coverage and ease of collection. However, several challenges arise in the capture of remote sensing images: first, objects are often very small relative to the size of the image, leading to limited feature information; second, objects can be oriented arbitrarily, requiring detection algorithms to learn rotation-invariant features for better performance; and third, noise and occlusions need to be handled by detection algorithms. In this paper, we propose MR-YOLOV5, a convolutional network specifically designed for real-time object detection in remote sensing images. MR-YOLOV5 is capable of detecting small objects with arbitrary orientations and tight spacing, as well as performing large-scale object detection. Additionally, by integrating multi-modal remote sensing data, the network’s performance in detecting challenging objects is enhanced. Experimental results on the publicly available VEDAI dataset demonstrate the effectiveness of the proposed algorithm.

关键词： Target recognition Surveillance Data integration Object detection Military aircraft Feature extraction Aircraft navigation Sensors detection algorithms Remote sensing

来源：评论

学校读者我要写书评

暂无评论

MS-UFAD: A Large-Scale Dataset for Real-world Unified Face Attack detection with Text Descriptions

MS-UFAD: A Large-Scale Dataset for Real-world Unified Face A...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Ning Jiang Dingheng Zeng Liang Gao Sheng Chen Zhifei Kong Yanhong Liu Jiao Li Yue Feng Tongtong Yuan Weihong Deng Quan Lu Ying Li School of Software & Microelectronics Peking University Beijing China Mashang Consumer Finance Co. Ltd. China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

As deepfake and adversarial attacks evolve, facial recognition systems are encountering increasingly diverse threats. Most existing face liveness detection algorithms focus on single tasks, like spoofing or deepfake attack detection. The corresponding datasets have limited coverage of attack methods, with original data mostly sourced from the internet or laboratory environments. Moreover, existing datasets lack textual annotations, particularly for attack clues, limiting algorithms’ ability to utilize semantic assistance from text. To address these issues, we propose a large-scale unified attack dataset, which includes newly collected facial videos from 5,000 individuals, along with generated videos corresponding to 52 face attack methods. The dataset contains 795k videos and 60k images across four different quality levels. Through semi-automated annotation, we provide detailed textual descriptions. This is the first face attack dataset with textual descriptions. Additionally, we propose a text-guided face attack detection method, demonstrating significant improvements in accuracy using fine-grained textual descriptions. Our dataset will be released at https://***.

关键词： Deepfakes Protocols Limiting Annotations Face recognition Semantics Laboratories Speech processing detection algorithms Faces

来源：评论

学校读者我要写书评

暂无评论

DeepFake-o-meter v2.0: An Open Platform for DeepFake detection

DeepFake-o-meter v2.0: An Open Platform for DeepFake Detecti...

引用

IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)

作者： Yan Ju Chengzhe Sun Shan Jia Shuwei Hou Zhaofeng Si Soumyya Kanti Datta Lipeng Ke Riky Zhou Anita Nikolich Siwei Lyu University at Buffalo State University of New York Buffalo USA Amazon Lab126 Sunnyvale California USA University of Illinois Urbana-Champaign Illinois USA

ISBN: (数字)9798350351422

ISBN: (纸本)9798350351439

Deepfakes, as AI-generated media, have increasingly threatened media integrity and personal privacy with realistic yet fake digital content. This work introduces an open-source and user-friendly online platform, DeepFake-O-Meter v2.0, that integrates state-of-the-art methods for detecting DeepFake images, videos, and audio. Built upon DeepFake-O-Meter v1.0, we have significantly upgraded and improved the platform architecture design, including user interaction, detector integration, job balancing, and security management. The platform aims to offer everyday users a convenient service for analyzing DeepFake media using multiple state-of-the-art detection algorithms. It ensures secure and private delivery of the analysis results. Furthermore, it serves as an evaluation and benchmanrking platform for researchers in digital media forensics to compare the performance of multiple algorithms on the same input. We have also conducted a detailed usage analysis based on the collected data to gain deeper insights into our platform's statistics. This involves analyzing four-month trends in user activity and evaluating the processing efficiency of each detector.

关键词： Performance evaluation Deepfakes Privacy Security management Forensics Detectors Information processing Media Market research detection algorithms

来源：评论

学校读者我要写书评

暂无评论

Detecting computer-generated disinformation

引用

INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS 2022年第4期13卷 363-383页

作者： Stiff, Harald Johansson, Fredrik Swedish Def Res Agcy FOI Stockholm Sweden

Modern neural language models can be used by malicious actors to automatically produce textual content looking as it has been written by genuine human users. Due to progress in the controllability of computer-generated text, there is a risk that state-sponsored actors may start using such methods for conducting large-scale information operations. Various detection algorithms have been suggested in the research literature to identify texts produced by language model-based generators, but these are often mainly evaluated on test data from the same distribution as they have been trained on. We evaluate promising Transformer-based detection algorithms in a large variety of experiments involving both in-distribution and out-of-distribution test data, as well as evaluation on more realistic in-the-wild data. It is shown that the generalizability of the detectors can be questioned, especially when applied to short social media posts. Moreover, the best performing (RoBERTa-based) detector is shown to be non-robust also to basic adversarial attacks, illustrating how easy it is for malicious actors to avoid detection by the current state-of-the-art detection algorithms.

关键词： Computer-generated text detection algorithms Information operations Language models

来源：评论

学校读者我要写书评

暂无评论

An Improved YOLO11-Based Algorithm for Small Object detection in UAV Aerial Imagery

An Improved YOLO11-Based Algorithm for Small Object Detectio...

引用

Artificial Intelligence, Networking and Information Technology (AINIT), International Seminar on

作者： Congming Wen Zhaotong Zhu Wenchao He Jing Wang School of Information Science and Engineering Yunnan University Kunming China Xi'an Precision Machinery Research Institute Kunming Branch China State Shipbuilding Corporation limited Kunming China

ISBN: (数字)9798331522285

ISBN: (纸本)9798331522292

Object detection in aerial imagery presents significant challenges due to the small size, low resolution, and indistinct features of targets. While YOLO-based algorithms are effective for detecting larger objects, they struggle with the detection of small targets. To mitigate this deficiency, an improved UAV detection method leveraging YOLO11 is presented. By incorporating dynamic convolutions in place of traditional ones, our approach improves both detection accuracy and computational efficiency. Furthermore, the integration of the PKI Module and CAA modules expands the receptive field and refines feature attention, while the SPPF-ECA module enhances channelspecific focus. A FFDPN structure is also introduced to improve multi-scale feature fusion. Results from experiments show that our approach exceeds the performance of YOLO and other cuttingedge models, achieving substantial gains in mean average precision (mAP) and small-object detection.

关键词： YOLO Seminars Accuracy Image resolution Autonomous aerial vehicles Feature extraction detection algorithms Vehicle dynamics Information technology Load modeling

来源：评论

学校读者我要写书评

暂无评论

A LightGBM Model for Liver Disease detection Against Syrian Refugees in Jordan

A LightGBM Model for Liver Disease Detection Against Syrian ...

引用

Computational Intelligence Approaches and Applications (ICCIAA), International Conference on

作者： Anas Khaleel Mahmou E. Farfoura Hakam Alqabani Jamal Hashem Department of Clinical Pharmacy and Pharmacy Practice University of Petra Amman Jordan Cybersecurity Department Al-Zaytoonah University of Jordan Amman Jordan Department of Pharmacy and Biomedical Sciences University of Strathclyde Glasgow United Kingdom Royal College of Surgeons in Ireland Medical University of Bahrain Adliya Kingdom of Bahrain

ISBN: (数字)9798331523657

ISBN: (纸本)9798331523664

The current report used LightGBM approach for liver function and disease detection among Syrian refugees in Jordan. The suggested LightGBM method integrates strong feature mining procedures with scalable classification structures for detection of liver function and disease. This report evaluates LightGBM detection algorithm to achieve optimal accuracy and interpretability. Results demonstrate a high detection accuracy across refugees -achieved accuracy 95%, with liver enzymes readings confusion matrix investigation emphasizing the strengths and weaknesses of the detection model. The LightGBM approach not only augments detection accuracy but also provides hands-on applicable method in resource-limited healthcare setting. This study helps in addressing important health challenges faced by refugee in northern Jordan and highlights the role of cutting-edge LightGBM techniques in improving liver function and disease diagnostics.

关键词： Enzymes Accuracy Liver diseases Computational modeling Medical services Machine learning Feature extraction detection algorithms Computational intelligence Diseases

来源：评论

学校读者我要写书评

暂无评论

Unsupervised domain adaptation in human detection for various human pose

Unsupervised domain adaptation in human detection for variou...

引用

International Image Processing, Applications and Systems Conference (IPAS)

作者： Benoit Lagadec Ekaterina Kostrykina Esifrance Le Cannet France Universite Cote d’Azur Nice Cedex 2 France

ISBN: (数字)9798331506520

ISBN: (纸本)9798331506537

In human detection algorithms, some recurrent issues are still challenging. Indeed, in a real application we have a lack of various human poses in our data. For example, in video surveillance it could be a significant issue to forget detecting a human in uncommon pose. We propose a compensation to these less frequent cases. We introduce a detection in rotated view to compensate limited various poses in our dataset. Instead of introducing a generative enhancement for every human body pose, which could be exhausting. Indeed, data annotation has become a difficulty for many applications. Moreover, a simple data augmentation is not efficient to generalize our training. We need a strategy to learn this new poses. So, we will introduce a process to compute pseudo labels in a sequence of rotated views. It avoids to use a clustering algorithm like dbscan [Chen et al.(2021)Chen, Lagadec, and Bremond.] or K-means to compute a pseudo label. A summary of theoretical results is described in a first section to illustrate our baseline. First of all, we will illustrate how we process pseudo label on various views by two mechanisms which have been called relaxation and scanning. We introduce a sequence of rotations of original view, to process a detection in a compensated view. In second step, we will learn this distortion created by the various views and construct an approach to distill original knowledge to a more generalized *** main improvement of this approach is to propose an alternative solution at dbscan [Chen et al.(2021)Chen, Lagadec, and Bremond.] approach which is popular, but which is difficult to use in real applications. We obtain a significant result by improving accuracy in especially various pose datasets.

关键词： Training Accuracy Annotations Image processing Clustering algorithms Data collection Video surveillance Distortion Data augmentation detection algorithms

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：