检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Nazemi, Amir Sepanj, Mohammad Hadi Pellegrino, Nicholas Czarnecki, Chris Fieguth, Paul Vision & Image Processing Lab Department of Systems Design Engineering University of Waterloo WaterlooON Canada

Current strategies for solving image-based inverse problems apply latent diffusion models to perform posterior sampling. However, almost all approaches make no explicit attempt to explore the solution space, instead drawing only a single sample from a Gaussian distribution from which to generate their solution. In this paper, we introduce a particle-filtering-based framework for a nonlinear exploration of the solution space in the initial stages of reverse SDE methods. Our proposed particle-filtering-based latent diffusion (PFLD) method and proposed problem formulation and framework can be applied to any diffusion-based solution for linear or nonlinear inverse problems. Our experimental results show that PFLD outperforms the SoTA solver PSLD on the FFHQ-1K and ImageNet-1K datasets on inverse problem tasks of super resolution, Gaussian debluring and inpainting. © 2024, CC BY-NC-ND.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Self-adjusting beat detection and prediction in music

Self-adjusting beat detection and prediction in music

引用

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

作者： R. Harper M.E. Jernigan Vision and Image Processing Lab Systems Design Engineering University of Waterloo Waterloo ONT Canada

This paper proposes a new approach to beat detection and prediction in music. Recurrent timing networks are used to detect and predict periodicities in an onset stream and are contained within nodes that compete for selection as the best beat hypothesis. Beat prediction nodes perform period self adjustment to better represent the detected music beat period. The system is tested using a variety of music from different genres and shows promise, in many cases with high correct beat detection percentages.

关键词： Multiple signal classification Timing Music Signal processing Delay Humans Intelligent networks Image processing design engineering systems engineering and theory

来源：评论

学校读者我要写书评

暂无评论

Memory-Efficient Continual Learning Object Segmentation for Long Videos

arXiv

引用

arXiv 2023年

作者： Nazemi, Amir Shafiee, Mohammad Javad Gharaee, Zahra Fieguth, Paul Vision & Image Processing Lab Department of Systems Design Engineering University of Waterloo WaterlooON Canada

Recent state-of-the-art semi-supervised Video Object Segmentation (VOS) methods have shown significant improvements in target object segmentation accuracy when information from preceding frames is used in segmenting the current frame. In particular, such memory-based approaches can help a model to more effectively handle appearance changes (representation drift) or occlusions. Ideally, for maximum performance, Online VOS methods would need all or most of the preceding frames (or their extracted information) to be stored in memory and be used for online learning in later frames. Such a solution is not feasible for long videos, as the required memory size grows without bound, and such methods can fail when memory is limited and a target object experiences repeated representation drifts throughout a video. We propose two novel techniques to reduce the memory requirement of Online VOS methods while improving modeling accuracy and generalization on long videos. Motivated by the success of continual learning techniques in preserving previously-learned knowledge, here we propose Gated-Regularizer Continual Learning (GRCL), which improves the performance of any Online VOS subject to limited memory, and a Reconstruction-based Memory Selection Continual Learning (RMSCL), which empowers Online VOS methods to efficiently benefit from stored information in memory. We also analyze the performance of a hybrid combination of the two proposed methods. Experimental results show that the proposed methods are able to improve the performance of Online VOS models by more than 8%, with improved robustness on long-video datasets while maintaining comparable performance on short-video datasets such as DAVIS16, DAVIS17, and YouTube-VOS18. © 2023, CC BY-NC-SA.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Comparing LBP, HOG and Deep Features for Classification of Histopathology Images

Comparing LBP, HOG and Deep Features for Classification of H...

引用

International Joint Conference on Neural Networks

作者： Taha J. Alhindi Shivam Kalra Ka Hin Ng Anika Afrin Hamid R. Tizhoosh Systems Design Engineering University of Waterloo Canada Electrical and Computer Engineering University of Waterloo Canada Kimia Lab University of Waterloo Canada

Medical image analysis has become a topic under the spotlight in recent years. There is a significant progress in medical image research concerning the usage of machine learning. However, there are still numerous questions and problems awaiting answers and solutions, respectively. In the present study, comparison of three classification models is conducted using features extracted using local binary patterns, the histogram of gradients, and a pre-trained deep network. Three common image classification methods, including support vector machines, decision trees, and artificial neural networks are used to classify feature vectors obtained by different feature extractors. We use KIMIA Path960, a publicly available dataset of 960 histopathology images extracted from 20 different tissue scans to test the accuracy of classification and feature extractions models used in the study, specifically for the histopathology images. SVM achieves the highest accuracy of 90.52% using local binary patterns as features which surpasses the accuracy obtained by deep features, namely 81.14%.

关键词： Feature extraction Support vector machines Machine learning Histograms Cancer Decision trees Neural networks

来源：评论

学校读者我要写书评

暂无评论

Skin lesion segmentation: U-Nets versus clustering

Skin lesion segmentation: U-Nets versus clustering

引用

IEEE Symposium Series on Computational Intelligence (SSCI)

作者： Bill S. Lin Kevin Michael Shivam Kalra H.R. Tizhoosh Department of Mechanical and Mechatronics University of Waterloo Waterloo Canada Department of Systems Design Engineering University of Waterloo Waterloo Canada KIMIA Lab University of Waterloo Waterloo Canada

Many automatic skin lesion diagnosis systems use segmentation as a preprocessing step to diagnose skin conditions because skin lesion shape, border irregularity, and size can influence the likelihood of malignancy. This paper presents, examines and compares two different approaches to skin lesion segmentation. The first approach uses U-Nets and introduces a histogram equalization based preprocessing step. The second approach is a C-Means clustering based approach that is much simpler to implement and faster to execute. The Jaccard Index between the algorithm output and hand segmented images by dermatologists is used to evaluate the proposed algorithms. While many recently proposed deep neural networks to segment skin lesions require a significant amount of computational power for training (i.e., computer with GPUs), the main objective of this paper is to present methods that can be used with only a CPU. This severely limits, for example, the number of training instances that can be presented to the U-Net. Comparing the two proposed algorithms, U-Nets achieved a significantly higher Jaccard Index compared to the clustering approach. Moreover, using the histogram equalization for preprocessing step significantly improved the U-Net segmentation results.

关键词： Skin Lesions Image segmentation Training Image color analysis Neural networks Hair

来源：评论

学校读者我要写书评

暂无评论

Skin Lesion Segmentation: U-Nets versus Clustering

arXiv

引用

arXiv 2017年

作者： Lin, Bill S. Michael, Kevin Kalra, Shivam Tizhoosh, H.R. Department of Mechanical and Mechatronics University of Waterloo Waterloo Canada Department of Systems Design Engineering University of Waterloo Waterloo Canada KIMIA Lab University of Waterloo Waterloo Canada

关键词： Dermatology

来源：评论

学校读者我要写书评

暂无评论

Semi-automatic Fisher-Tippett Guided Active Contour for Lumbar Multifidus Muscle Segmentation

Semi-automatic Fisher-Tippett Guided Active Contour for Lumb...

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society

作者： Dorothy Lui Christian Scharfenberger Diana E. De Carvalho Jack P. Callaghan Alexander Wong Faculty of Systems Design Engineering Vision and Image Processing Lab University of Waterloo Waterloo Canada The Department of Kinesiology University of Waterloo Waterloo Canada

Rehabilitative Ultrasound Imaging or diagnostic ultrasound is used to measure geometric properties of the lumbar multifidus muscle to infer muscle strength or degeneration for back pain therapy. For this purpose, a novel semi-automatic approach (FTS: Fisher-Tippett Segmentation) based upon the Decoupled Active Contour is proposed to reliably and quickly segment the lumbar multifidus muscle in diagnostic ultrasound. To overcome speckle or hardly visible region boundaries in ultrasound images, we first propose a novel external energy functional to explicitly consider the underlying Fisher-Tippett distribution of ultrasound data. We then introduce a user-guided Hidden Markov Model trellis formation for improved segmentation of weakly-defined regions. Extensive experiments have shown that our approach not only improves the segmentation performance when compared to existing methods, but also does not rely on sub-specialized knowledge for segmentation.

关键词： Muscles Ultrasonic imaging Image segmentation Hidden Markov models Back Medical treatment Active contours

来源：评论

学校读者我要写书评

暂无评论

Stochastic Nonlinear Model Predictive Control of Battery-Supercapacitor Hybrid Energy Storage systems in Electric Vehicles

Stochastic Nonlinear Model Predictive Control of Battery-Sup...

引用

American Control Conference

作者： Parisa Golchoubian Nasser L. Azad Kumaraswamy Ponnambalam Smart Hybrid and Electric Vehicle Systems (SHEVS) lab in the Department of System Design Engineering University of Waterloo Department of System Design Engineering University of Waterloo

ISBN: (纸本)9781509045839

In this paper, a stochastic nonlinear model predictive control (S-NMPC) approach has been illustrated for the energy management of a battery/SC hybrid energy storage systems (HESSs) in a Toyota Rav4EV. As the performance of predictive controllers in these systems is highly dependent on the future power demand estimation, a stochastic approach has been adopted in order to include uncertainties in the driver's power demand prediction, exploiting the ideas of the two-stage stochastic programming method. The power demand is modeled as a Markov chain and has been trained with several standard and real-world driving cycles. Model-in-the-Loop (MIL) simulation results have been presented over a driving cycle different from that used as training data. The results of the S-NMPC method has been compared against other deterministic approaches.

关键词： performance predictive controllers Controllers Electric vehicles nonlinear model predictive control Power demand Predictive hybrid energy storage Energy management Markov chain Driving cycle

来源：评论

学校读者我要写书评

暂无评论

Opposition-Based Learning: A New Scheme for Machine Intelligence

Opposition-Based Learning: A New Scheme for Machine Intellig...

引用

International Conference on Computational Intelligence for Modelling, Control and Automation, and International Conference on Intelligent Agents, Web Technologies and Internet Commerce

作者： H.R. Tizhoosh Pattern Analysis and Machine Intelligence Lab Systems Design Engineering University of Waterloo Waterloo ONT Canada

Opposition-based learning as a new scheme for machine intelligence is introduced. Estimates and counter-estimates, weights and opposite weights, and actions versus counter-actions are the foundation of this new approach. Examples are provided. Possibilities for extensions of existing learning algorithms are discussed. Preliminary results are provided

关键词： Machine learning Machine intelligence Genetic algorithms Intelligent agent Neural networks Internet Biological neural networks Humans Learning systems Computational intelligence

来源：评论

学校读者我要写书评

暂无评论

Fast fuzzy edge detection

Fast fuzzy edge detection

引用

Conference of the North American Fuzzy Information Processing Society - NAFIPS

作者： H.R. Tizhoosh Pattern Recognition and Machine Intelligence Lab Systems Design Engineering University of Waterloo Waterloo ONT Canada

In the literature, there already exist some fuzzy approaches to edge detection. However, they are generally computationally expensive. In this paper, several fast fuzzy edge detectors are introduced for practical case... 详细信息

关键词： Image edge detection Detectors Robustness Fuzzy sets Pixel Fuzzy set theory Pattern recognition Machine intelligence design engineering systems engineering and theory

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：