检索结果-内蒙古大学图书馆

4th IEEE International Conference on Data Science and Computer Application, ICDSCA 2024

作者： Xia, Shiji Du, Yi Zhang, Jingyi Yang, Zuopeng Wang, Yu Fu, Yue Liu, Ming Ding, Xie School of Computer and Information Engineering Shanghai Polytechnic University Shanghai China Data Intelligence Product Department Wonders Information Co. Ltd Shanghai China

ISBN: (纸本)9798350368239

In the context of depth images, fusing human segmentation images with depth background images provides critical data for human detection tasks, where accurate localization of the human head is a key indicator. However, in real-world scenarios with occlusions and complex backgrounds, human segmentation images often lack head pixels. This missing data leads to subpar performance in human recognition tasks for deep learning models. To address this issue, this paper proposes an efficient approach that combines edge detection, projection analysis, and template matching for upper-body localization and head identification. Additionally, a refined diffusion model is used to repair the human head edge mask obtained from the initial processing. The main advantage of this approach is its independence from complex deep learning model training. It integrates traditional image processing techniques, morphological analysis, and diffusion models, simplifying head localization and effectively avoiding interference from the shoulder and arm regions. This approach enables more accurate head localization and uses diffusion models to restore missing portions of the human head, thereby improving overall pose recognition accuracy. Experimental results demonstrate that the repaired depth images outperform the original images in human pose model training. When trained with the repaired images, detection accuracy (mAPVal50-95) and precision improved by 10.1% and 7.1%, respectively, significantly enhancing downstream object detection performance. These findings validate the effectiveness of the proposed method in repairing missing human head depth images. They further highlight its potential in improving pose recognition and object detection performance, offering reliable technical support for related tasks. ©2024 IEEE.

关键词： Template matching

来源：评论

学校读者我要写书评

暂无评论

Wavelet-like Transform with Subbands Fusion in Decoupled Structure for deep image Compression

Wavelet-like Transform with Subbands Fusion in Decoupled Str...

引用

Picture Coding Symposium (PCS)

作者： Ma, Ke Wu, Yaojun Zhang, Zhaobin Esenlik, Semih Sun, Xiaoyan Zhang, Kai Zhang, Li Univ Sci & Technol China Hefei 230026 Anhui Peoples R China Bytedance Inc San Diego CA 92122 USA

ISBN: (纸本)9798350358483;9798350358490

Wavelet-like transform, based on convolutional neural network (CNN), is content-adaptive and has made remarkable achievements in end-to-end image compression. However, the subsequent sequential processing of each subband in the entropy module takes a relatively long decoding time, resulting in inconvenience for real-world applications. In this work, for lossy image compression, the wavelet-like transform is transplanted into the prevailing autoencoder structure to enhance the analysis and synthesis transform due to its excellent decomposition capability. The obtained subbands of different frequencies will undergo a hierarchical decorrelation architecture for subband fusion, also called cross fusing module. The specialized treatment will be applied to different subbands according to their spatial resolution to attain a more compact latent representation. In addition, the proposed solution features an architecture that decouples the arithmetic decoding process from the sample prediction process, which significantly reduces the decoding complexity. Experiments on the Kodak test set show that the proposed method achieves -3.04% BD-Rate compared to existing decoupled end-to-end structure in RGB Peak Signal-to-Noise Ratio (PSNR).

关键词： neural image compression wavelet transform decoupled entropy coding hierarchical decorrelation

来源：评论

学校读者我要写书评

暂无评论

An Efficient Diffusion for Blind Face Restoration 24

An Efficient Diffusion for Blind Face Restoration

引用

7th International Conference on image and Graphics processing, ICIGP 2024

作者： Wang, Yu Wang, Gencheng Chen, Rong Chen, Yuzhen Xizang Minzu University Xian Yang China

ISBN: (纸本)9798400716720

In recent years, many deep learning-based methods especially the diffusion model have made a remarkable breakthrough in blind face restoration. Thanks to the powerful generative capabilities of the diffusion model, these methods can generate relatively better results when confronted with intricate degradation without multiple constraints. However, the inference efficiency of diffusion models is hampered by the iterative sampling process inherent in model inference. To address this problem, we focus on the lightweight and fast diffusion model. In this study, we propose an efficient diffusion for blind face restoration (EDBFR). The key of our method is to introduce a specialized diffusion model solver which crafted to refine the diffusion partial differential equations to enhance the sampling speed. Additionally, we devise a lightweight residual module and integrate it into the Unet network of diffusion models to reduce the number of parameters while maintaining performance. Comprehensive experiments on CelebA-Test, LFW-Test and WebPhoto-Test demonstrate that our proposed method surpasses the state-of-the-art deep learning-based blind face restoration methods both quantitatively and qualitatively. Especially, compared with the state-of-the-art diffusion models, our model has 20% inference time, and 50% parameters. © 2024 ACM.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Research on Precision Machining Error Compensation Algorithm Based on Machine learning

Research on Precision Machining Error Compensation Algorithm...

引用

2024 International Conference on Electronics and Devices, Computational Science, ICEDCS 2024

作者： Xiaoyu, Huan Zhenjiang Branch of Jiangsu Union Technical Institute Mechanical and Electrical Engineering Department Zhenjiang212016 China

ISBN: (纸本)9798331527624

In the field of precision machining, especially for mechanical thin-walled blades with complex geometric shapes, the precise control of machining errors is the key to achieve high quality products and improve production efficiency. In this paper, an error compensation algorithm based on machine learning is proposed, which aims to significantly reduce the errors in precision machining process through intelligent prediction and real-time correction. The algorithm design integrates deep learning model and traditional control theory to automatically learn and optimize compensation strategies based on historical processing data. The system simulation experiments were carried out in a virtual machining environment, covering the blade machining process with different materials and machining conditions. The experimental results show that the algorithm can reduce the machining error by 30% on average, and the error compensation rate is as high as 45% under certain conditions. In addition, the nonlinear relationship between machining parameters and error compensation effect is revealed, which provides a theoretical basis for optimizing machining parameters. © 2024 IEEE.

关键词： Quality control

来源：评论

学校读者我要写书评

暂无评论

Research on power ecological data service model based on privacy computing and computing power network technology

Research on power ecological data service model based on pri...

引用

2024 International Conference on Cloud Computing, Performance Computing, and deep learning, CCPCDL 2024

作者： Song, Yinghua Fan, Qiaoyun Nanjing211106 China NARI Technology Co. Ltd. Nanjing211106 China China Mobile Group Jiangsu Co. Ltd. Jiangsu Nanjing210006 China

ISBN: (纸本)9781510683044

Data has become an important factor of production, and the collection, storage and value sharing of data have become increasingly important. At present, there is a difficult problem of how to realize data value mining and sharing on the premise of ensuring data security. In order to solve these problems, this article conducts relevant research on the key technologies of the "data+cloud" power ecological data service model based on privacy computing and computing power network technology, and proposes to use privacy computing technology and computing power network to build a "data+cloud" power ecological data service realizes multi-party data joint modeling and analysis, thereby building a safer and more reliable data application solution. © 2024 SPIE.

关键词： Data modeling Data privacy Data processing Distributed computing Computer security Clouds Industry Network security Network architectures real-time computing

来源：评论

学校读者我要写书评

暂无评论

A deep Automated System for Detection of Fake Currency 2

A Deep Automated System for Detection of Fake Currency

引用

2nd International Conference on Inventive Computing and Informatics (ICICI)

作者： Sree, Kota Sai Suneetha, S. Giribabu, S. Pace Inst Technol Sci Dept Comp Sci & Engn Ongole AP India

ISBN: (纸本)9798350373301;9798350373295

Generating fake currency becomes more complicated for the country's economy. This shows a significant impact on economic growth and harm to national security. Many existing models are available to find fake currency by using machine learning (ML) and deep learning (DL) algorithms. The existing models have several issues detecting fake currency, such as more processing time, highly expensive, and mismatched results. The proposed approach is called the automated approach (AA), which finds the accurate fake currency. In this study, the proposed automated approach combines RESNET50 for training and Generative Adversarial Networks (GANs) to detect and locate fake currency effectively. The training RESNET50 collects the fake currency images and obtains the patterns and factors showing the difference between original and fake currency. The preprocessing technique of Bilateral Filtering is used to denoise the image without losing the edges of the photos. The Grayscale Conversion (GC) model is used as the feature extraction technique, which shows the significant impact on the final output. The proposed GAN model shows high performance in detecting fake currency. Results show that the automated approach performed better in detecting and classifying fake currency.

关键词： Fake Currency Convolutional Neural Network (CNN) Feature Extraction

来源：评论

学校读者我要写书评

暂无评论

Automated classification of pancreatic neuroendocrine tumors using label-free multiphoton microscopy and deep learning

Automated classification of pancreatic neuroendocrine tumors...

引用

Conference on Label-Free Biomedical Imaging and Sensing (LBIS)

作者： Guan, Shuyuan Daigle, Noelle Sawyer, Travis W. Univ Arizona Wyant Coll Opt Sci 1630 E Univ Blvd Tucson AZ 85721 USA Univ Arizona Dept Biomed Engn 1127 E James E Rogers Way Tucson AZ 85721 USA Univ Arizona Coll Med 1515 N Campbell Ave Tucson AZ 85724 USA

ISBN: (纸本)9781510669680;9781510669673

Pancreatic neuroendocrine tumors (PNETs) present significant diagnostic and therapeutic challenges due to their heterogeneity and complex nature as a subtype of pancreatic cancer. The treatment approach varies considerably based on the tumor's location, grading, and focality. Accurate prognosis and management typically necessitate the expertise of a pathologist to evaluate histological slides of the tissue, a process that is often time-consuming and labor-intensive. Developing point-of-care techniques for automatic classification of PNETs would greatly improve the ability to treat and manage this disease by providing real-time decision-making information. In response to these challenges, our study introduces a highly efficient and versatile diagnostic strategy. This innovative approach synergistically integrates label-free multiphoton microscopy with finely adjusted, pre-trained deep learning models, optimized for performance even with limited data availability. We have meticulously optimized four pre-trained convolutional neural networks, utilizing a dataset comprising only 49 images, which includes both two-photon excitation fluorescence and second-harmonic generation imaging. This refined approach has resulted in an impressive average classification accuracy of over 95% for the development dataset and more than 90% for the test dataset. These results are significantly superior when compared to the preoperative misdiagnosis rates of conventional diagnostic modalities such as ultrasound (US) and computed tomography (CT), which stand at 81.8% and 61.5%, respectively. This methodology represents a significant advancement in the diagnostic process for PNETs, promising a more streamlined, rapid, and accurate approach to treatment. Furthermore, it opens substantial potential for the automated classification of various tumor types using multiphoton microscopic imaging, even in scenarios characterized by limited data availability.

关键词： pancreatic neuroendocrine tumor multiphoton microscopy deep learning convolutional neural networks image classification

来源：评论

学校读者我要写书评

暂无评论

End-to-end speaker recognition system design 12

End-to-end speaker recognition system design

引用

SPIE 12th International Symposium on Multispectral image processing and Pattern Recognition, MIPPR 2023

作者： Wang, Liheng Ji, Xinran Liao, Shengjie Zhang, Hailong School of Electrical Information Wuhan Institute of Technology East Lake New Technology Development Zone Hubei Wuhan430205 China Huazhong University of Science and Technology No. 1037 Luoyu Road Hongshan District Hubei Wuhan430074 China

ISBN: (数字)9781510675001

ISBN: (纸本)9781510674998

Like fingerprints, facial information and other biological characteristics, human voice also carries the physiological characteristics of living things. It is unique, stable personal information that cannot be stolen or lost. The speaker's voiceprint can contain changes that remain unchanged, so these features make the voiceprint features deeper, more elusive, and more difficult to forge, making the authentication stronger and safer. Based on the basic characteristics of human voiceprints, this paper designs a voiceprint recognition system based on neural networks. In order to enable the network to better use original data to obtain output results, time and frequency domain masking methods are used for data enhancement. In the network part, the encoder-decoder method uses the transformer architecture to achieve end-to-end data processing, and uses the triplet loss function to evaluate and optimize the parameters within the neural network to improve the prediction accuracy of the model. Modeling experiments were conducted on the LibriSpeech and CN-Celeb datasets, respectively. The system realizes human voiceprint recognition end-to-end based on deep learning and has been tested to meet the design needs. © 2024 SPIE.

关键词： deep learning

来源：评论

学校读者我要写书评

暂无评论

Assisted documentation as a new focus for artificial intelligence in endoscopy: the precedent of reliable withdrawal time and image reporting

引用

ENDOSCOPY 2023年第12期55卷 1118-1123页

作者： Lux, Thomas J. Sassmannshausen, Zita Kafetzis, Ioannis Sodmann, Philipp Herold, Katja Sudarevic, Boban Schmitz, Ruediger Zoller, Wolfram G. Meining, Alexander Hann, Alexander Univ Hosp Wurzburg Internal Med 2 Intervent & Expt Endoscopy InExEn Wurzburg Germany Katharinen Hosp Dept Internal Med & Gastroenterol Stuttgart Germany Univ Hosp Hamburg Eppendorf Dept Interdisciplinary Endoscopy D-20246 Hamburg Germany Univ Hosp Hamburg Eppendorf Dept Internal Medicine1 Hamburg Germany Univ Hosp Hamburg Eppendorf Dept Computat Neurosci Hamburg Germany

Background Reliable documentation is essential for maintaining quality standards in endoscopy;however, in clinical practice, report quality varies. We developed an artificial intelligence (AI)-base d prototype for the measurement of withdrawal and intervention times, and automatic *** A multiclass deep learning algorithm distinguishing different endoscopic image content was trained with 10557 images (1300 examinations, nine centers, four processors). Consecutively, the algorithm was used to calculate withdrawal time (AI prediction) and extract relevant images. Validation was performed on 100 colonoscopy videos (five centers). The reported and AI-predicted withdrawal times were compared with video-bas ed measurement;photodocumentation was compared for documented *** Video-based measurement in 100 colonoscopies revealed a median absolute difference of 2.0 minutes between the measured and reported withdrawal times, compared with 0.4 minutes for AI predictions. The original photodocumentation represented the cecum in 88 examinations compared with 98/100 examinations for the AI-generated documentation. For 39/104 polypectomies, the examiners' photographs included the instrument, compared with 68 for the AI images. Lastly, we demonstrated real-time capability (10 colonoscopies).Conclusion Our AI system calculates withdrawal time, provides an image report, and is real-time ready. After fur-ther validation, the system may improve standardized re-port ing, while decreasing the workload created by routine documentation.

关键词： Colorectal cancer CRC screening documentation image and data processing Quality management

来源：评论

学校读者我要写书评

暂无评论

Efficient Flower Classification Using CNN with InceptionV3 Transfer learning 12

Efficient Flower Classification Using CNN with InceptionV3 T...

引用

12th International Conference on Information and Communication Technology, ICoICT 2024

作者： Anggara, Fiky Da Assuncao Dias Ximenes, Jubileu Sulistiyo, Mahmud Dwi Saputra, Muhamad Fikry Hadiyoso, Sugondo School of Computing Telkom University Bandung Indonesia School of Applied Science Telkom University Bandung Indonesia

ISBN: (纸本)9798350363432

Flowers are attractive due to their unique colors and shapes, so it is important to classify the uniqueness of flower species through observation. The classification of various flower species is crucial for identifying new or rare species, contributing to medicinal advancements, and understanding their role in feeding insects, birds, animals, and humans. Automatic classification of flower species can be done by utilizing image-based technology, which has become a highlight for experts to develop digital flower catalogs. However, a number of studies on flower species recognition have been conducted, many of which focus on traditional approaches to image processing. Currently, research is increasingly leading to the application of deep learning as well as transfer learning in image-based object recognition tasks. In this research, the InceptionV3 transfer learning architecture on the Convolutional Neural Network (CNN) is implemented for flower species classification. The dataset used consists of five types of flower species images, namely daisy, dandelion, rose, sunflower, and tulip from the flower dataset. The experimental results show that the developed model achieves an accuracy rate of 94% for train data and 88.9% for test data for CNN InceptionV3. The average inference time for these images was 0.073639 seconds, this indicates that the model is highly efficient in processing and classifying flower images. The correct classification results for flower species were 93.29% for daisy, 99.97% for dandelion, 99.89% for rose, 92.47% for sunflower, and 55.57% for tulip. These findings demonstrate the high effectiveness of the CNN InceptionV3 model-based architecture for classifying flower species. © 2024 IEEE.

关键词： Invertebrates

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：