检索结果-内蒙古大学图书馆

2007 ieee symposium on computational intelligence in image and signal processing, CIISP 2007

ISBN: (纸本)1424407079

The proceedings contain 76 papers. The topics discussed include: a new vehicle detection approach in traffic jam conditions;non-pixel robot stereo;a novel method to recognize complex dynamic gesture by combining HMM and FNN models;boundary refined texture segmentation based on K-views and datagram methods;single-row superposition-type spherical compound-like eye for pan-tilt motion recovery;bare bones strategy for human detection and tracking;Daubechies complex wavelet transform based moving object tracking;identification of dynamic nonlinear systems using computational intelligence techniques;a new invariant descriptor for shape representation and recognition;a wavelet-fuzzy logic based system to detect and identify electric disturbs;evolution strategies based particle filters for fault detection;and a multi-window stereo vision algorithm with improved performance at object borders.

关键词： Tracking (position)

来源：评论

学校读者我要写书评

暂无评论

Generating image Counterfactuals in Deep Learning Models Without the Aid of Generative Models

引用

ieee signal processing LETTERS 2025年 32卷 1495-1499页

作者： Xu, Ao Li, Zihao Zhang, Yukai Wu, Tieru Jilin Univ Sch Artificial Intelligence Changchun 130000 Peoples R China

With the rapid development of artificial intelligence, particularly the rise of deep learning, the importance of Explainable Artificial intelligence has become increasingly prominent. Among its key techniques, counterfactual explanation plays a crucial role in understanding the decision-making mechanisms of opaque models. However, the high dimensionality and complex feature patterns of image data pose significant challenges for the task of generating counterfactuals for images. Existing literature has proposed various algorithms based on different assumptions, many of which rely on the existence of appropriate generative models. Some of these assumptions, particularly the assumption regarding the existence of generative models, may be overly stringent. To address this issue, this letter introduces a novel assumption-free image counterfactual generation algorithm, DFO-S, based on Score Matching and gradient-free optimization techniques. The proposed method achieves high-quality counterfactual generation without relying on generative models. Through extensive empirical analysis, we demonstrate the significant superiority of our method in terms of performance.

关键词： signal processing algorithms Artificial intelligence Closed box Deep learning Vectors Optimization Data models computational modeling Training Smoothing methods explainable artificial intelli- gence image counterfactual explanation

来源：评论

学校读者我要写书评

暂无评论

Adaptive Bit Selection for Scalable Deep Hashing

引用

ieee TRANSACTIONS ON image processing 2025年 34卷 1048-1059页

作者： Wang, Min Zhou, Wengang Yao, Xin Li, Houqiang Hefei Comprehens Natl Sci Ctr Inst Artificial Intelligence Hefei 230000 Peoples R China Univ Sci & Technol China Dept Elect Engn & Informat Sci MoE Key Lab Brain Inspired Intelligent Percept & C Hefei 230027 Peoples R China

Deep Hashing is one of the most important methods for generating compact feature representation in content-based image retrieval. However, in various application scenarios, it requires training different models with diversified memory and computational resource costs. To address this problem, in this paper, we propose a new scalable deep hashing framework, which aims to generate binary codes with different code lengths by adaptive bit selection. Specifically, the proposed framework consists of two alternative steps, i.e., bit pool generation and adaptive bit selection. In the first step, a deep feature extraction model is trained to output binary codes by optimizing retrieval performance and bit properties. In the second step, we select informative bits from the generated bit pool with reinforcement learning algorithm, in which the same retrieval performance and bit properties are directly used in computing reward. The bit pool can be further updated by fine-tuning the deep feature extraction model with more attention on the selected bits. Hence, these two steps are alternatively iterated until convergence is achieved. Notably, most existing binary hashing methods can be readily integrated into our framework to generate scalable binary codes. Experiments on four public image datasets prove the effectiveness of the proposed framework for image retrieval tasks.

关键词： Quantization (signal) Binary codes Costs Measurement Feature extraction computational modeling Databases Representation learning Training Hash functions Adaptive bit selection deep hashing scalable hashing image retrieval

来源：评论

学校读者我要写书评

暂无评论

A Novel Clutter Suppression Algorithm for Airborne Radar via Subspace Projection

引用

ieee SENSORS JOURNAL 2025年第1期25卷 962-970页

作者： Zhang, Xinying Wang, Tong Cui, Weichen Wang, Degen Xidian Univ Natl Key Lab Radar Signal Proc Xian 710071 Peoples R China Beijing Aerosp Automatic Control Inst Natl Key Lab Sci & Technol Aerosp Intelligence Con Beijing 100854 Peoples R China Southwest China Res Inst Elect Equipment Chengdu 610036 Peoples R China

Strong ground clutter submerges the moving targets and hinders the radar image analysis. An effective technique called space-time adaptive processing (STAP) is used to suppress clutter in airborne radar systems but has the disadvantage of large training sample requirements. Sparse recovery STAP (SR-STAP) has the benefit of improving the algorithm performance when the training sample size is small. However, the computational burden and storage space restrict the practical application of SR-STAP. To address this issue, a fast subspace-based SR-STAP algorithm is proposed in this article. Using the Bayesian framework, the proposed algorithm selects the most matched atoms to span a clutter subspace by minimizing the likelihood function. In comparison with the existing SR-STAP algorithm, the proposed algorithm avoids the time-consuming sparse reconstruction process. Moreover, the estimated subspace can be applied to gain-phase error calibration. The proposed algorithm is experimentally demonstrated to have superior performances in terms of clutter suppression, target detection, and computational complexity.

关键词： Clutter Vectors Noise Training Sensors Airborne radar signal processing algorithms Object detection Bayes methods Atoms Clutter suppression gain-phase errors space-time adaptive processing (STAP) sparse recovery subspace estimated

来源：评论

学校读者我要写书评

暂无评论

IRAU-Net: Inception Residual Attention U-Net in Adversarial Network for Cardiac MRI Segmentation

引用

ieee JOURNAL OF SELECTED TOPICS IN signal processing 2025年第1期19卷 260-272页

作者： Rostami, Maryam Talebi Motamedi, Seyed Ahmad Amirkabir Univ Technol Elect Engn Dept Tehran *** Iran

Due to the significant advancements in medical imaging, the application of artificial intelligence for early disease diagnosis has greatly contributed to reducing mortality caused by heart diseases. Cardiac MRI segmentation into the left ventricle (LV), right ventricle (RV) and myocardium (MYO) is a key step towards the early treatment and diagnosis of heart diseases. The proposed method consists of two main components: a localization part to extract the heart organ from other parts of the image, and a GAN model for segmentation. In the generator part of the generative adversarial network (GAN) model, a customized U-Net network is employed to segment the cardiac image. A combination of inception and residual blocks and utilizing an attention mechanism yield an improved version of U-Net. On the other hand, the discriminator part of the GAN model is designed to accurately distinguish between the ground truth image and the segmented image generated by the generator part. Our method is evaluated on two cardiac MRI datasets. The evaluation results on the ACDC 2017 challenge dataset show mean Dice scores of 0.947 for LV, 0.919 for RV, and 0.907 for MYO in cardiac MRI segmentation. The experimental results highlight that our proposed IRAU-Net method outperforms other state-of-the-art methods in terms of accuracy while significantly reducing computational costs.

关键词： image segmentation Generators Generative adversarial networks Heart Accuracy Magnetic resonance imaging Location awareness Detectors Computer architecture Semantic segmentation Attention mechanism cardiac segmentation deep learning generative adversarial network medical image segmentation U-Net

来源：评论

学校读者我要写书评

暂无评论

DiffHSR: Unleashing Diffusion Priors in Hyperspectral image Super-Resolution

引用

ieee signal processing LETTERS 2025年 32卷 236-240页

作者： Jia, Yizhen Xie, Yumeng An, Ping Tian, Zhen Hua, Xia Shanghai Univ Sch Commun & Informat Engn Shanghai 200444 Peoples R China Guangdong Lab Artificial Intelligence & Digital Ec Shenzhen 518000 Peoples R China

Hyperspectral images provide rich spectral information and have been widely applied in numerous computer vision tasks. However, their low spatial resolution often limits their use in applications such as image segmentation and recognition. In previous works, generating high-resolution hyperspectral (HR-HS) images required the use of low-resolution hyperspectral (LR-HS) images and high-resolution RGB (HR-RGB) images as priors, which increases the cost of data collection and may lead to measurement and calibration errors in practical applications. Although the currently popular CNN-based single hyperspectral image super-resolution (single HS-SR) methods have improved performance, they are not flexible enough to process images with different degradation. From a visual perspective, the generated super-resolution images exhibit a significant smudging effect due to the loss of information. Leveraging multi-modal techniques and generative prior, we propose DiffHSR that marks a significant leap in LR-HS images super-restoration without HR-RGB. Additionally, we have established a connection between hyperspectral images and the RGB image-based generative model tasks using low-cost data and fine-tuning approaches, which creates a novel paradigm. Comprehensive experiments have demonstrated that our proposed method achieves strong visual performance and competitive results in term of quantitative metrics and perceptive quality.

关键词： Hyperspectral imaging Superresolution Degradation Visualization Lighting image reconstruction Diffusion models computational modeling Adaptation models Three-dimensional displays Generative prior multi-modal techniques single hyperspectral image super-resolution

来源：评论

学校读者我要写书评

暂无评论

From Body Parts to Holistic Action: A Fine-Grained Teacher-Student CLIP for Action Recognition

引用

ieee signal processing LETTERS 2025年 32卷 1336-1340页

作者： Ou, Yangjun Shi, Xiao Chen, Jia He, Ruhan Liu, Chi Wuhan Text Univ Sch Comp Sci & Artificial Intelligence Wuhan 430200 Peoples R China TCL Corp Res Wuhan 430000 Peoples R China

Action recognition in dynamic video remains challenging, particularly when distinguishing between visually similar actions. While existing methods often rely on holistic representations, they overlook the fine-grained details that are significant for accurate classification. We propose a novel Fine-grained Teacher-student CLIP (FT-CLIP) that integrates body part analysis with holistic action recognition through a teacher-student architecture, bridging the gap between fine-grained action parsing and overall action understanding. The teacher model processes individual body parts alongside specialized description to generate part-specific features, which are then aggregated and distilled into the student model. Through knowledge distillation with learnable prompts, the student model effectively learns to capture subtle action distinctions while maintaining efficient inference. FT-CLIP achieves a more nuanced understanding of complex actions by progressing from detailed body part analysis to comprehensive action recognition. Experiments on Kinetics-TPS under a fully-supervised setting and on HMDB51 and UCF101 under a zero-shot setting demonstrate the effectiveness of our method.

关键词： Training Accuracy computational modeling Motion segmentation image segmentation image recognition Artificial intelligence Visualization Textiles Head Action recognition vision-language learning knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

MANTIS: A Mixed-signal Near-Sensor Convolutional imager SoC Using Charge-Domain 4b-Weighted 5-to-84-TOPS/W MAC Operations for Feature Extraction and Region-of-Interest Detection

引用

ieee JOURNAL OF SOLID-STATE CIRCUITS 2025年第3期60卷 934-948页

作者： Lefebvre, Martin Bol, David Catholic Univ Louvain Inst Informat & Commun Technol Elect & Appl Math B-1348 Louvain La Neuve Belgium Delft Univ Technol Dept Microelect NL-2628 CD Delft Netherlands

Recent advances in artificial intelligence (AI) have prompted the search for enhanced algorithms and hardware to support the deployment of machine learning (ML) at the edge. More specifically, in the context of the Internet of Things (IoT), vision chips must be able to fulfill the tasks of low to medium complexity, such as feature extraction (FE) or region-of-interest (RoI) detection, with a sub-mW power budget imposed by the use of small batteries or energy harvesting. Mixed-signal vision chips relying on in-or near-sensor processing have emerged as an interesting candidate because of their favorable tradeoff between energy efficiency (EE) and computational accuracy compared with digital systems for these specific tasks. In this article, we introduce a mixed-signal convolutional imager system-on-chip (SoC) codenamed MANTIS, featuring a unique combination of large 16 x 16 4b-weighted filters, operation at multiple scales, and double sampling, well suited to the requirements of medium-complexity tasks. The main contributions are (i) circuits called DS3 units combining delta-reset sampling (DRS), image downsampling (DS), and voltage downshifting and (ii) charge-domain multiply-and-accumulate (MAC) operations based on switched-capacitor (SC) amplifiers and charge sharing in the capacitive DAC of the successive-approximation (SAR) ADCs, MANTIS achieves peak EEs normalized to 1b operations of 4.6 and 84.1 TOPS/W at the accelerator and SoC levels, while computing feature maps (fmaps) with a root-mean-square error (RMSE) ranging from 3 to 11.3%. It also demonstrates a face RoI detection with a false negative rate (FNR) of 11.5%, while discarding 81.3% of image patches and reducing the data transmitted off chip by 13 x compared with the raw image.

关键词： Charge domain CMOS image sensor (CIS) convolutional neural network (CNN) feature extraction (FE) mixed signal multiply-and-accumulate (MAC) operations near sensor region-of-interest (RoI) detection system-on-chip (SoC) Charge domain CMOS image sensor (CIS) convolutional neural network (CNN) feature extraction (FE) mixed signal multiply-and-accumulate (MAC) operations near sensor region-of-interest (RoI) detection system-on-chip (SoC)

来源：评论

学校读者我要写书评

暂无评论

Language-Enhanced Dual-Level Contrastive Learning Network for Open-Set Hyperspectral image Classification

引用

ieee TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2025年 63卷

作者： Qin, Boao Feng, Shou Zhao, Chunhui Li, Wei Tao, Ran Zhou, Jun Harbin Engn Univ Minist Ind & Informat Technol Coll Informat & Commun Engn Harbin 150001 Peoples R China Harbin Engn Univ Minist Ind & Informat Technol Key Lab Adv Marine Commun & Informat Technol Harbin 150001 Peoples R China Minist Nat Resources Technol Innovat Ctr South China Sea Remote Sensing Guangzhou 510300 Peoples R China Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China Griffith Univ Sch Informat & Commun Technol Nathan Qld 4111 Australia

In recent years, language-supervised vision models have demonstrated impressive potential in learning open-world concepts. Some research has introduced this learning paradigm to the hyperspectral image (HSI) processing domain;however, there has been limited work integrating textual information into the hyperspectral open-set recognition task. To fill this gap, we leverage textual supervision information in open-set HSI classification (HSIC) and propose a language-enhanced dual-level contrastive learning network (LDCLNet). Specifically, we introduce a linguistic mode with prior knowledge as a supervised signal to enhance the metric distances between closed-set samples and provide supplementary semantic information for open-set samples. Second, a dual-level visual-language (V-L) contrastive learning (CL) approach, which can align visual and language embeddings separately at the instance level and manifold level, is proposed to establish a more accurate link between visual and language representations. Finally, a distance-refined open-set recognition method is proposed, which aims to effectively discover unknown class samples during testing by refining predictions of known and unknown classes. Extensive experiments and analysis on three public HSI datasets validate the effectiveness of LDCLNet.

关键词： Visualization Training Testing Feature extraction Hyperspectral imaging Contrastive learning Artificial intelligence Semantics image classification computational modeling Dual-level contrastive learning (CL) hyperspectral image (HSI) language-supervised vision model open-set classification

来源：评论

学校读者我要写书评

暂无评论

Latent Weight Quantization for Integerized Training of Deep Neural Networks

引用

ieee TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE intelligence 2025年第4期47卷 2816-2832页

作者： Fei, Wen Dai, Wenrui Zhang, Liang Zhang, Luoming Li, Chenglin Zou, Junni Xiong, Hongkai Shanghai Jiao Tong Univ Dept Elect Engn Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ Dept Comp Sci & Engn Shanghai 200240 Peoples R China Donghua Univ Sch Comp Sci & Technol Shanghai 201620 Peoples R China Zhejiang Univ Key Lab Biomed Engn Minist Educ Hangzhou 310027 Peoples R China

Existing methods for integerized training speed up deep learning by using low-bitwidth integerized weights, activations, gradients, and optimizer buffers. However, they overlook the issue of full-precision latent weights, which consume excessive memory to accumulate gradient-based updates for optimizing the integerized weights. In this paper, we propose the first latent weight quantization schema for general integerized training, which minimizes quantization perturbation to training process via residual quantization with optimized dual quantizer. We leverage residual quantization to eliminate the correlation between latent weight and integerized weight for suppressing quantization noise. We further propose dual quantizer with optimal nonuniform codebook to avoid frozen weight and ensure statistically unbiased training trajectory as full-precision latent weight. The codebook is optimized to minimize the disturbance on weight update under importance guidance and achieved with a three-segment polyline approximation for hardware-friendly implementation. Extensive experiments show that the proposed schema allows integerized training with lowest 4-bit latent weight for various architectures including ResNets, MobileNetV2, and Transformers, and yields negligible performance loss in image classification and text generation. Furthermore, we successfully fine-tune Large Language Models with up to 13 billion parameters on one single GPU using the proposed schema.

关键词： Quantization (signal) Training Perturbation methods Memory management Hardware Trajectory Random access memory Graphics processing units computational modeling Noise Integerized training deep neural network quantization latent weight dual quantizer large language models

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：