检索结果-内蒙古大学图书馆

2023 Wave Electronics and its Application in Information and Telecommunication systems, WECONF 2023

作者： Sivov, N. Yu. Shmatko, E.v. Poroykov, A. Yu. National Research University MPEI Department of Physics Named after V.A. Fabrikant Moscow Russia

ISBN: (纸本)9798350348293

Computer image simulation is used in the development of optical measurement systems and processing algorithms. It allows the performance of a system or algorithm to be verified and its measurement uncertainty to be assessed. Modern algorithms make it possible to achieve sub-pixel accuracy in digital image processing. An example of such algorithms is the cross-correlation processing in Particle image velocimetry. Often, standard graphics engines such as Unity 3D and Unreal Engine are used to simulate the images. The purpose of this work is to evaluate the possibility of using these tools to generate images for further processing by subpixel resolution algorithms. This paper presents the results of a computational experiment to estimate the maximum spatial resolution for simulations in Unity 3D. A qualitative and quantitative analysis of the results confirms the possibility of using the tested graphics engine for the analysis of subpixel resolution algorithms for digital image processing. © 2023 IEEE.

关键词： Engines

来源：评论

学校读者我要写书评

暂无评论

An adaptive radial object recognition algorithm for lightweight drones in different environments

COMPUTER OPTICS

引用

COMPUTER OPTICS 2025年第3期49卷 480-492页

作者： Song, S. Liu, J. Shleimovich, M. P. Shakirzyanov, R. M. Novikova, S. v. Yang Zhou Marine Elect Instruments Inst Yangzhou 225101 Jangsu Peoples R China Kazan Natl Res Tech Univ st K Marksa 10 Kazan 420111 Russia

The paper proposes a group of radial shape object recognition methods capable of finding many different-sized circular objects in an image with high accuracy in minimum time and conditions of uneven brightness of frame areas. The methods are not computationally demanding, making them suitable for use in computer vision systems of light unmanned vehicles, which cannot carry powerful computing devices on board. The methods are also suitable for unmanned vehicles traveling at high speed, where image processing must be performed in real-time. The proposed algorithms are robust to noise. When combined into a single group, the developed algorithms constitute a customizable set capable of adapting to different imaging conditions and computing power. This property allows the method to be used for detecting objects of interest in different environments: from the air, from the ground, underwater, and when moving the vehicle between these environments. We proposed three methods: a hybrid FRODAS method combines the FRST and Hough methods to increase accuracy and reduce the time to search for circles in the image;a PaRCIS method based on sequential image compression and reconstruction to increase the speed of searching for multiple circles of different radii and removing noise;an additional modification of LIPIS is used with any of the primary or developed methods to reduce the sensitivity to sharp variations in the frame's brightness. The paper presents comparative experiments demonstrating the advantages of the developed methods over classical circle recognition methods regarding accuracy and speed. It shows the advantage of recognizing circles of different brightness. Experiments on recognizing multiple real-world objects in photographs taken on the ground, in the air, and underwater, with complex scenes under distortion and blurring with different degrees of illumination, demonstrate the effectiveness of the set of methods.

关键词： degrees computer vision multiple object recognition image compression recognition within a sliding window non-uniform image brightness changing shooting conditions

来源：评论

学校读者我要写书评

暂无评论

Feature Map Guided Adapter Network for Object Detection in Low-light Conditions

Feature Map Guided Adapter Network for Object Detection in L...

引用

IEEE International Symposium on Circuits and systems (ISCAS)

作者： Pang, Cong Zhou, Wei Li, Haoyan Zhang, Xiangyu Lou, Xin ShanghaiTech Univ Sch Informat Sci & Technol Shanghai Peoples R China Minist Educ Key Lab Intelligent Percept & Human Machine Colla Shanghai Peoples R China

ISBN: (纸本)9798350330991;9798350331004

Conventional ISP pipelines and image enhancement methods are designed and optimized for human vision, creating a gap between the requirements of computer and human visions. To bridge the requirement gap, we present a co-design framework in which backend computer vision plays a pivotal role in shaping the proceeding image processing algorithm. It features a pre-processing adapter network, responsible for the restoration and enhancement of RAW images from computer vision perspective, especially in challenging environmental conditions. Specifically, we extract feature maps from the backend vision network, utilizing them as constraints for optimizing the preprocessing adapter network. To validate the effectiveness of our proposed framework, we employ object detection in low-light conditions as the computer vision task, with YOLO-v5 as the backbone. Given the considerable noise in low-light images, we compare our results with state-of-the-art denoising algorithms, showcasing the superior performance of our framework.

关键词： Computer vision image processing co-design raw image perceptual loss

来源：评论

学校读者我要写书评

暂无评论

Mitotic cell detection in histopathological images of neuroendocrine tumors using improved YOLOv5 by transformer mechanism

引用

SIGNAL image AND vIDEO processing 2023年第8期17卷 4107-4114页

作者： Yucel, Zehra Akal, Fuat Oltulu, Pembe Necmettin Erbakan Univ Dept Comp Technol Konya Turkiye Hacettepe Univ Grad Sch Sci & Engn Ankara Turkiye Hacettepe Univ Fac Engn Dept Comp Engn Ankara Turkiye Necmettin Erbakan Univ Fac Med Dept Pathol Konya Turkiye

Automatic analysis of pathological images is important for the diagnosis and treatment of diseases. The use of computerized systems in this field is becoming increasingly common. Due to evolving technology and the speed of information needed, it is desirable for computers to be able to recognize objects like humans. Deep learning methods, which are a subfield of artificial intelligence, and image processing algorithms that recognize objects from images have been used in many fields in recent years, including healthcare. The aim of this study is to detect the mitoses in the histopathological images of neuroendocrine tumors using image processing methods based on deep learning. In our study, You Only Look Once-v5 (YOLOv5), the most widely used object recognition method, was used by combining the YOLOv5 transform module. YOLOv5 recognized mitotic cells with an accuracy of 0.80, a recall of 0.67, and an F1 score of 0.73, while the YOLOv5 transformer model recognized mitotic cells with an accuracy of 0.89, a recall of 0.68, and an F1 score of 0.77. The acceleration of the process and the objective evaluation will contribute significantly to an accurate and fast diagnosis. Another advantage is the time saved for pathologists, who can concentrate on important cases. In summary, automatic mitotic cell detection will facilitate tumor grade determination, treatment, and patient monitoring.

关键词： Biomedical image processing Histopathological images Mitosis detection YOLOv5 Transformer

来源：评论

学校读者我要写书评

暂无评论

S²P: Two-Stage Superpixel Algorithm for Enhanced Lane Detection on Resource Constraint Edges 14

S<SUP>2</SUP>P: Two-Stage Superpixel Algorithm for Enhanced ...

引用

14th International Symposium on Industrial Embedded systems

作者： Rhuthik, P. vaddhiparthi, S. v. S. L. N. Surya Suhas Kannan, Pranav Gangadharan, Deepak Int Inst Informat Technol Hyderabad India

ISBN: (纸本)9798350388640;9798350388633

Lane marker identification is crucial for developing Intelligent Transportation systems and Autonomous vehicles. While deep learning models excel in accuracy, their high computational requirements make them unsuitable for low-power edge devices. While various image-processing-based algorithms are used to pre-process images at different stages, there is a lack of an efficient algorithm specifically designed to enhance the lane-modeling stage. This paper proposes a two-stage pre-processing algorithm, consisting of a static phase and a dynamic phase, to implement a divide-and-conquer approach for enhancing existing lane-modeling algorithms. The static phase generates an initial pixel label map and seed locations, while the dynamic phase uses the pre-processed input image and seed locations to generate a dynamic pixel map for localizing lane marking zones. Focusing on the center two lanes, the proposed method improves overall accuracy and recall, which is critical for capturing smaller lane details. Experimental results demonstrate the superior performance of the proposed two-stage algorithm compared to traditional HT and PHT methods, with the two-stage algorithm outperforming the standard HT(0.88 vs 0.92) and PHT(0.89 vs 0.95) in terms of Average Accuracy and Recall. Additionally, the two-stage PHT method significantly reduces power consumption compared to the Ultra Fast Lane Detection (UFLD) model, making it ideal for low-power edge devices like the Raspberry Pi 4B and Jetson Nano.

关键词： Intelligent Transportation systems Autonomous vehicles image processing Lane Modeling Hough Transform

来源：评论

学校读者我要写书评

暂无评论

Real-Time Dehazing and Defogging: A Comprehensive Analysis for Single image Haze and Fog Removal

Real-Time Dehazing and Defogging: A Comprehensive Analysis f...

引用

2024 IEEE International Conference on Information Technology, Electronics and Intelligent Communication systems, ICITEICS 2024

作者： Kumari, Apurva Kumar, A. Pavan Teja, B.v.S. Tarun B v Raju Institute of Technology Department of Electronics and Communication Engineering Telangana India

ISBN: (纸本)9798350382693

Implementing image dehazing and defogging on a Field Programmable Gate Array (FPGA) offers efficiency. Dehazing an image becomes particularly challenging in the presence of fog or haze. However, employing a dark channel prior to dehazing allows the removal of haze particles from the image. Different modules are used in this process, such as Dynamic Atmospheric Light Estimation (DALE), Scene Recovery (SR), and Transmission Map Estimation (TME). The FPGA runs these modules in hardware and produces effective outcomes by employing image processing algorithms. In this case, using FPGA technology offers a number of benefits. The dehazing process can be accelerated by using FPGA's built-in parallel processing capabilities to execute numerous operations at once. Furthermore, FPGA implementations provide better throughput and reduced latency in comparison to conventional approaches, making them well-suited for real-time applications such as image dehazing and D © 2024 IEEE.

关键词： image analysis

来源：评论

学校读者我要写书评

暂无评论

Latin Square and Machine Learning Techniques Combined Algorithm for image Encryption

引用

CIRCUITS systems AND SIGNAL processing 2023年第11期42卷 6829-6853页

作者： Patel, Sakshi Thanikaiselvan, v. Vellore Inst Technol VIT Sch Elect Engn Vellore 632014 India

Multimedia data is crucial in the military, medical, forensics, social, etc., to transmit a large amount of data. Security of this sensitive information is the primary issue. This paper uses Latin square and machine learning techniques such as neural networks and genetic algorithm to design an image encryption algorithm. A new neural network-based pseudorandom number generator is proposed to generate a chaotic sequence for various applications. Encryption key images are designed using Latin squares in the finite field. Further, the Latin squares are XOR with the input matrix to get the encrypted images. The proposed algorithm is iterated a finite number of times to generate a cipher image population. Randomly two parents are chosen from the generated population, and row and column arrangements produce offspring. A genetic algorithm is the optimization technique used for the best encrypted image search. The pixel correlation value serves as a fitness function. Finally, the least correlated cipher image is obtained from the genetic algorithm applied to the parent and offspring of the population generated from the encryption algorithm. The simulation results from the proposed image encryption model surpass many communication channel attacks and perform better when compared to existing image security algorithms.

关键词： image encryption Neural network Pseudorandom number generator Latin square Genetic algorithm Machine learning techniques

来源：评论

学校读者我要写书评

暂无评论

Modified deep inductive transfer learning diagnostic systems for diabetic retinopathy severity levels classification

引用

BIOMEDICAL SIGNAL processing AND CONTROL 2025年 99卷

作者： vij, Richa Arora, Sakshi Shri Mata Vaishno Devi Univ Sch Comp Sci & Engn Katra 182320 Jammu & Kashmir India

Diabetic Retinopathy (DR), a retinal illness that degenerates the retina and causes blindness, can be effectively treated with early detection and examination. Although expensive and unpleasant, manual retinography is the gold standard for DR diagnosis. Many Deep Learning (DL)-based algorithms have shown promise as deep learning (DR) diagnostic tools, performing similarly to human picture evaluation. These strategies work best with finetuned hyperparameters and huge databases. To develop Deep Inductive Transfer Learning-based Diagnostic systems (DITL-DS) for Multiclass DR Severity Classification. The benchmark Indian Diabetic Retinopathy image Dataset (IDRiD) is chosen first. Next, the dataset is pre-processed by proportionally scaling the image to 128*128 and using Convolutional Local Area Histogram Equalization (CLAHE). Third, the uneven class label dataset is balanced for efficiency. Fourth, a global average pooling layer and bottlenecking are added to five DITL models-Inception v3, ResNet34, EfficientNet B0, vGG16, and Xception-to compensate for information loss. Finally, we compare our improved approaches to five basic DITL classifiers. Our modified Xception model separates severity stages best, with an accuracy of 0.9840, precision of 0.9845, recall of 0.9650, AUC of 0.9979, and AUC-ROC of 0.9902. Precision, recall, and F1 scores are 0.97, 0.99, and 0.98 for stage0, 0.99, 1, and 1 for stage1, 0.98, 0.94, and 0.96 for stage2, 0.99, 0.99, and 0.99 for stage3, and 1, 1, and 1 for stage4. This may help clinicians objectively diagnose DR early. To ensure robustness, the model's generalizability was validated using APTOS (Asia Pacific Tele-Ophthalmology Society).

关键词： Diabetic retinopathy Deep learning Deep inductive transfer learning-based diagnostic systems Multiclass DR severity classification Global average pooling layer IDRiD dataset IDRiD dataset

来源：评论

学校读者我要写书评

暂无评论

Comparison of Brain Tumor Detection Techniques by Using Different Machine Learning YOLO algorithms 1

引用

4th Congress on Intelligent systems (CIS)

作者： Tasnim, Faria Islam, Md Tobibul Maisha, Aniqa Tahsin Sultana, Israt Akter, Tasnia Islam, Md Toufiqul Mil Inst Sci & Technol Dhaka Bangladesh M Abdur Rahim Med Coll Dinajpur Bangladesh

ISBN: (数字)9789819990405

ISBN: (纸本)9789819990399;9789819990405

Cell division that is out of control and aberrant leads to brain tumors. They can be developed in the brain as well as in the tissue of the lymphatic system, vessels of blood, nerves of the cranial, and brain envelopes. Furthermore, brain tumors may enlarge due to the metastasis of cancers primarily located in other body sections. Brain tumors can endanger patients' lives, including physical, emotional, and financial risks. Hence, early diagnosis of a brain tumor is vital to lead to successful treatment and recovery. Imaging tests involving computerized tomography (CT) and or magnetic resonance imaging (MRI) tend to be used to make a diagnosis of brain tumors. However, it is challenging to automatically detect a brain tumor in its early stages using conventional image processing techniques. Therefore, several algorithms combined with conventional image processing methods for brain tumors have been developed over time. In this paper, pretrained YOLO v4 (You Only Look Once) and YOLO v7 models have been used for a comparative analysis of the real-time detection of brain tumors based on preprocessing techniques as well as detection units. MRI images have been collected from Kaggle and have been preprocessed with Gaussian, average, and median filters integrating with contrast limited adaptive histogram equalization and distinct datasets that have been created. Each set of images was annotated using Roboflow to create test, train, and validate images. The brain tumor is then detected from MRI images using pretrained YOLO models. In current research work, the precision is about 96% for the dataset which has been preprocessed by Gaussian filter integrating with CLAHE and trained on model YOLO v7, which is relatively high compared to the model YOLO v7. The current method can be used by physicians in real-time brain tumor detection, advancing our healthcare system.

关键词： Brain tumors YOLO v4 YOLO v7 Magnetic resonance imaging Computerized tomography

来源：评论

学校读者我要写书评

暂无评论

High-Dimensional Data processing Using Quantum-Inspired Evolutionary algorithms for Homeland Security Imaging systems

High-Dimensional Data Processing Using Quantum-Inspired Evol...

引用

2024 International Conference on Communication, Computer Sciences and Engineering, IC3SE 2024

作者： Alekhya, v. Reddy, N. v. Uma Singh, Jay Boddu, Bhasker Sobti, Rajeev Hameed, Ali Abduhussien Institute Of Aeronautical Engineering Dundigal Hyderabad India New Horizon College Of Engineering Department Of Artificial Intelligence And Machine Learning Bangalore India Gl Bajaj Institute Of Technology & Management U.P. Greater Noida India Mlr Institute Of Technology Department Of Information Technology Telangana Hyderabad India Lovely Professional University Phagwara India College Of Medical Technology The Islamic University Radiology Techniques Department Najaf Iraq

ISBN: (纸本)9798350366846

This research study explores the emerging area of quantum-inspired evolutionary algorithms (QIEAs) applied to high-dimensional data processing, with a focus on homeland security imaging systems. This work attempts to close the paradigm gap in image processing methodologies caused by the growing complexity of security threats. By combining the adaptive processes of evolutionary algorithms with the probabilistic reasoning inherent in quantum computing, the proposed technique creates a potent tool for managing the inherent difficulties of high-dimensional data. The main contribution of this study is the creation of a brand-new QIEA framework that is specifically designed for image analysis in security contexts and exhibits higher accuracy and efficiency than traditional techniques. The quantum bit representation and quantum gate operations, which have been cleverly tailored to the evolutionary algorithm structure, provide the methodological basis, and improve the search capacity in multidimensional data fields. An unparalleled degree of clarity and detail in security photography is made possible by this fusion, which is essential for threat identification and prevention. Experiments conducted on a variety of difficult datasets show that the suggested method is reliable for identifying important characteristics in complicated photos, which is a crucial component of homeland security applications. Beyond only addressing short-term security issues, this study establishes a standard for next investigations into quantum-inspired computing for image processing. © 2024 IEEE.

关键词： Imaging systems

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：