检索结果-内蒙古大学图书馆

A Fast Attention-Guided Hierarchical Decoding network for Real-Time Semantic Segmentation

SENSORS 2024年第1期24卷 95页

作者： Hu, Xuegang Feng, Jing Chongqing Univ Posts & Telecommun Sch Commun & Informat Engn Chongqing 400065 Peoples R China Chongqing Univ Posts & Telecommun Chongqing Key Lab Signal & Informat Proc Chongqing 400065 Peoples R China

Semantic segmentation provides accurate scene understanding and decision support for many applications. However, many models strive for high accuracy by adopting complex structures, decreasing the inference speed, and making it challenging to meet real-time requirements. Therefore, a fast attention-guided hierarchical decoding network for real-time semantic segmentation (FAHDNet), which is an asymmetric U-shaped structure, is proposed to address this issue. In the encoder, we design a multi-scale bottleneck residual unit (MBRU), which combines the attention mechanism and decomposition convolution to design a parallel structure for aggregating multi-scale information, making the network perform better at processing information at different scales. In addition, we propose a spatial information compensation (SIC) module that effectively uses the original input to make up for the spatial texture information lost during downsampling. In the decoder, the global attention (GA) module is used to process the feature map of the encoder, enhance the feature interaction in the channel and spatial dimensions, and enhance the ability to mine feature information. At the same time, the lightweight hierarchical decoder integrates multi-scale features to better adapt to different scale targets and accurately segment objects of different sizes. Through experiments, FAHDNet performs outstandingly on two public datasets, Cityscapes and Camvid. Specifically, the network achieves 70.6% mean intersection over union (mIoU) at 135 frames per second (FPS) on Cityscapes and 67.2% mIoU at 335 FPS on Camvid. Compared to the existing networks, our model maintains accuracy while achieving faster inference speeds, thus enhancing its practical usability.

关键词： real-time semantic segmentation attention mechanism encoder-decoder network feature fusion

来源：评论

学校读者我要写书评

暂无评论

RRNet: Repetition-Reduction network for Energy Efficient Depth Estimation

引用

IEEE ACCESS 2020年 8卷 106097-106108页

作者： Oh, Sangyun Kim, Hye-Jin S. Lee, Jongeun Kim, Junmo Ulsan Natl Inst Sci & Technol UNIST Sch Elect & Comp Engn Ulsan 44776 South Korea Seoul Natl Univ Neural Proc Res Ctr Seoul 151742 South Korea Korea Adv Inst Sci & Technol Sch Elect Engn Robot Program Daejeon 34141 South Korea ETRI Artificial Intelligence Lab Daejeon 34129 South Korea

Lightweight neural networks that employ depthwise convolution have a significant computational advantage over those that use standard convolution because they involve fewer parameters;however, they also require more time, even with graphics processing units (GPUs). We propose a Repetition-Reduction network (RRNet) in which the number of depthwise channels is large enough to reduce computation time while simultaneously being small enough to reduce GPU latency. RRNet also reduces power consumption and memory usage, not only in the encoder but also in the residual connections to the decoder. We apply RRNet to the problem of resource-constrained depth estimation, where it proves to be significantly more efficient than other methods in terms of energy consumption, memory usage, and computation. It has two key modules: the Repetition-Reduction (RR) block, which is a set of repeated lightweight convolutions that can be used for feature extraction in the encoder, and the Condensed Decoding Connection (CDC), which can replace the skip connection, delivering features to the decoder while significantly reducing the channel depth of the decoder layers. Experimental results on the KITTI dataset show that RRNet consumes less energy and less memory than conventional schemes, and that it is faster on a commercial mobile GPU without increasing the demand on hardware resources relative to the baseline network. Furthermore, RRNet outperforms state-of-the-art lightweight models such as MobileNets, PyDNet, DiCENet, DABNet, and EfficientNet.

关键词： Computer vision deep neural network depth estimation encoder-decoder network lightweight neural network machine learning mobile graphical processing unit (GPU) unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

A Video Deblurring Algorithm Based on Motion Vector and An Encorder-decoder network

引用

IEEE ACCESS 2019年 7卷 86778-86788页

作者： Zhang, Shanqing Li, Pengcheng Meng, Yiheng Li, Li Zhou, Qili Fu, Xiaofeng Hangzhou Dianzi Univ Sch Comp Sci & Technol Hangzhou 310018 Zhejiang Peoples R China

Camera shakes cause video motion blur. Video deblurring has been studied for years, and however, there are still unresolved problems, such as video frame alignment, frame selection, and frame ambiguity evaluation. We propose a video deblurring algorithm based on the motion vector and an encoder-decoder network. Our algorithm consists of four steps: first, the blurry image blocks in a video frame are located using a blurred image quality evaluation algorithm based on a response function of singular values. Second, the corresponding candidates of the blurry image block in the consecutive frames are searched using the motion vector, and the optimal candidate blocks are obtained using an objective function. Third, the blurry image block and the optimal candidate blocks are served as samples, which are inputted to an encoder-decoder network, so that the blurry image block is repaired. Finally, all blurry image blocks are replaced with the repaired ones, the boundary artifacts are eliminated, and the entire video frame is repaired. The experiments show that our algorithm yields sharper repair results, and the overall performance of our algorithm is better than other related algorithms.

关键词： Video deblurring blur quality evaluation motion vector encoder-decoder network

来源：评论

学校读者我要写书评

暂无评论

PMSA-DyTr: Prior-Modulated and Semantic-Aligned Dynamic Transformer for Strip Steel Defect Detection

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2024年第4期20卷 6684-6695页

作者： Su, Jiaojiao Luo, Qiwu Yang, Chunhua Gui, Weihua Silven, Olli Liu, Li Cent South Univ Changsha 410083 Peoples R China Cent South Univ Sch Automat Changsha 410083 Peoples R China Univ Oulu Oulu 90014 Finland Natl Univ Def Technol Changsha 410073 Peoples R China

In-process hot-rolled strip steel is suffering from some complicated yet unavoidable surface defects due to its harsh production environment. The automated visual inspection on defects consistently faces challenges of interclass similarity, intraclass difference, low contrast, and overlapping issue, which tend to trigger false or missed detections. This article proposes a prior-modulated and semantic-aligned dynamic transformer, called PMSA-DyTr. In this framework, a long short-term self-attention embedded with local convolution is designed for assisting an encoder to eliminate noise ambiguity between defects and backgrounds. Then, a semantic aligner is cleverly bridged between the encoder and the decoder to align the sematic for speeding up the convergence, and prior-modulated cross attention is proposed to alleviate the deficiency of samples for a data-driven transformer. Furthermore, a gate controller is innovatively constructed to dynamically select the minimal number of encoder blocks while preserving detection accuracy. The proposed PMSA-DyTr outperforms 19 state-of-the-art models on mean average precision with an inference time of 54.67 ms and visually performs best in detecting low-contrast and multiple small defects.

关键词： Automated visual inspection (AVI) encoder-decoder network salient detection steel strip transformer

来源：评论

学校读者我要写书评

暂无评论

FDRN: A fast deformable registration network for medical images

引用

MEDICAL PHYSICS 2021年第10期48卷 6453-6463页

作者： Sun, Kaicong Simon, Sven Univ Stuttgart Inst Parallel & Distributed Syst Stuttgart Germany

Purpose Deformable image registration is a fundamental task in medical imaging. Due to the large computational complexity of deformable registration of volumetric images, conventional iterative methods usually face the tradeoff between the registration accuracy and the computation time in practice. In order to boost the performance of deformable registration in both accuracy and runtime, we propose a fast unsupervised convolutional neural network for deformable image registration. Methods The proposed registration model FDRN possesses a compact encoder-decoder network architecture which employs a pair of fixed and moving images as input and outputs a three-dimensional displacement vector field (DVF) describing the offsets between the corresponding voxels in the fixed and moving images. In order to efficiently utilize the memory resources and enlarge the model capacity, we adopt additive forwarding instead of channel concatenation and deepen the network in each encoder and decoder stage. To facilitate the learning efficiency, we leverage skip connection within the encoder and decoder stages to enable residual learning and employ an auxiliary loss at the bottom layer with lowest resolution to involve deep supervision. Particularly, the low-resolution auxiliary loss is weighted by an exponentially decayed parameter during the training phase. In conjunction with the main loss in high-resolution grid, a coarse-to-fine learning strategy is achieved. Last but not least, we involve a proposed multi-label segmentation loss (SL) to improve the network performance in Dice score in case the segmentation prior is available. Comparing to the SL using average Dice score, the proposed SL does not require additional memory in the training phase and improves the registration accuracy efficiently. Results We evaluated FDRN on multiple brain MRI datasets from different aspects including registration accuracy, model generalizability, and model analysis. Experimental results demonstrate

关键词： brain MRI registration coarse-to-fine learning deep supervision deformable image registration encoder-decoder network multi-label segmentation loss

来源：评论

学校读者我要写书评

暂无评论

Depth-Wise Asymmetric Bottleneck With Point-Wise Aggregation decoder for Real-Time Semantic Segmentation in Urban Scenes

引用

IEEE ACCESS 2020年 8卷 27495-27506页

作者： Li, Gen Jiang, Shenlu Yun, Inyong Kim, Jonghyun Kim, Joongkyu Sungkyunkwan Univ Dept Elect Elect & Comp Engn Suwon 16419 South Korea

Semantic segmentation is a process of linking each pixel in an image to a class label, and is widely used in the field of autonomous vehicles and robotics. Although deep learning methods have already made great progress for semantic segmentation, they either achieve great results with numerous parameters or design lightweight models but heavily sacrifice the segmentation accuracy. Because of the strict requirements of real-world applications, it is critical to design an effective real-time model with both competitive segmentation accuracy and small model capacity. In this paper, we propose a lightweight network named DABNet, which employs Depth-wise Asymmetric Bottleneck (DAB) and Point-wise Aggregation decoder (PAD) module to tackle the challenging real-time semantic segmentation in urban scenes. Specifically, the DAB module creates a sufficient receptive field and densely utilizes the contextual information, and the PAD module aggregates the feature maps of different scales to optimize performance through the attention mechanism. Compared with existing methods, our network substantially reduces the number of parameters but still achieves high accuracy with real-time inference ability. Extensive ablation experiments on two challenging urban scene datasets (Cityscapes and CamVid) have proved the effectiveness of the proposed approach in real-time semantic segmentation.

关键词： Real-time semantic segmentation encoder-decoder network convolutional neural network urban scenes lightweight network

来源：评论

学校读者我要写书评

暂无评论

Single infrared image enhancement using a deep convolutional neural network

引用

NEUROCOMPUTING 2019年 332卷 119-128页

作者： Kuang, Xiaodong Sui, Xiubao Liu, Yuan Chen, Qian Gu, Guohua Nanjing Univ Sci & Technol Nanjing 210094 Jiangsu Peoples R China

In this paper, we propose a deep learning method for single infrared image enhancement. A fully convolutional neural network (CNN) is used to produce images with enhanced contrast and details. The conditional generative adversarial networks are incorporated into the optimization framework to avoid the background noise being amplified and further enhance the contrast and details. The existing convolutional neural network architectures, such as residual architectures and encoder-decoder architectures, fail to achieve the best results both in terms of network performance and application scope for infrared image enhancement task. To address this problem, we specifically design a new refined convolutional neural architecture that produces visually very appealing results with higher contrast and sharper details compared to other network architectures. Visible images are used for training since there are fewer infrared images. Proper training samples are generated to ensure that the network trained on visible images can be well applied to infrared images. Experiments demonstrate that our approach outperforms existing image enhancement algorithms in terms of contrast and detail enhancement. Code is available at https://***/Kuangxd/IE-CGAN. (C) 2018 Elsevier B.V. All rights reserved.

关键词： Infrared images Enhancement Deep learning encoder-decoder network Residual network Generative adversarial network

来源：评论

学校读者我要写书评

暂无评论

RA-Net: Region-Aware Attention network for Skin Lesion Segmentation

引用

COGNITIVE COMPUTATION 2024年第5期16卷 2279-2296页

作者： Naveed, Asim Naqvi, Syed S. Iqbal, Shahzaib Razzak, Imran Khan, Haroon Ahmed Khan, Tariq M. COMSATS Univ Islamabad CUI Dept Elect & Comp Engn Islamabad 45550 Pakistan Univ Engn & Technol UET Lahore Dept Comp Sci & Engn Narowal Campus Narowal 51600 Pakistan UNSW Sch Comp Sci & Engn Sydney NSW 2052 Australia

The precise segmentation of skin lesion in dermoscopic images is essential for the early detection of skin cancer. However, the irregular shapes of the lesions, the absence of sharp edges, the existence of artifacts like hair follicles, and marker color make this task difficult. Currently, fully connected networks (FCNs) and U-Nets are the most commonly used techniques for melanoma segmentation. However, as the depth of these neural network models increases, they become prone to various challenges. The most pertinent of these challenges are the vanishing gradient problem and the parameter redundancy problem. These can result in a decline in Jaccard index of the segmentation model. This study introduces a novel end-to-end trainable network designed for skin lesion segmentation. The proposed methodology consists of an encoder-decoder, a region-aware attention approach, and guided loss function. The trainable parameters are reduced using depth-wise separable convolution, and the attention features are refined using a guided loss, resulting in a high Jaccard index. We assessed the effectiveness of our proposed RA-Net on four frequently utilized benchmark datasets for skin lesion segmentation: ISIC 2016, ISIC 2017, ISIC 2018, and PH2. The empirical results validate that our method achieves state-of-the-art performance, as indicated by a notably high Jaccard index.

关键词： Attention approach encoder-decoder network Skin lesion segmentation Convolutional neural network Cancer diagnosis

来源：评论

学校读者我要写书评

暂无评论

Discriminative Feature network Based on a Hierarchical Attention Mechanism for Semantic Hippocampus Segmentation

引用

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2021年第2期25卷 504-513页

作者： Shi, Jiali Zhang, Rong Guo, Lijun Gao, Linlin Ma, Huifang Wang, Jianhua Ningbo Univ Fac Elect Engn & Comp Sci Ningbo 315000 Zhejiang Peoples R China Northwest Normal Univ Coll Comp Sci & Engn Lanzhou 730000 Gansu Peoples R China Ningbo Univ Dept Radiol Affiliated Hosp Med Sch Ningbo 315000 Zhejiang Peoples R China

The morphological analysis of hippocampus is vital to various neurological studies including brain disorders and brain anatomy. To assist doctors in analyzing the shape and volume of the hippocampus, an accurate and automatic hippocampus segmentation method is highly demanded in the clinical practice. Given that fully convolutional networks (FCNs) have made significant contributions in biomedical image segmentation applications, we propose a notably discriminative feature network based on a hierarchical attention mechanism in hippocampal segmentation. First, considering the problem that the hippocampus is a rather small part in MR images, we design a contexta-ware high-level feature extraction module (CHFEM) to extract high-level features of scale invariance in the encoder stage. Further, we introduce a hierarchical attention mechanism into our segmentation framework. The mechanism is divided into three parts: a low-level feature spatial attention module (LFSAM) is developed to learn the spatial relationship between different pixels on each channel in the low-level stage of the encoder, a high-level feature channel attention module (HFCAM) is to model the semantic information relationship on different channel images in the high-level stage of the encoder, and a cross-connected attention module (CCAM) is designed in the decoder part to further suppress the noisy boundaries of hippocampus and simultaneously utilize the attentional low-level features from the encoder to better guide the high-level hippocampus edge segmentation in the decoder phase. The proposed approach achieves outstanding performance on the ADNI dataset and the Decathlon dataset compared with other semantic segmentation models and existing hippocampal segmentation approaches. Source code is available at https:***'Hippocampal-segmentation.

关键词： Hippocampal segmentation encoder-decoder network high-level features low-level features

来源：评论

学校读者我要写书评

暂无评论

CGAN: closure-guided attention network for salient object detection

引用

VISUAL COMPUTER 2022年第11期38卷 3803-3817页

作者： Das, Dibyendu Kumar Shit, Sahadeb Ray, Dip Narayan Majumder, Somajyoti CSIR CMERI AcSIR Durgapur W Bengal India CSIR CMERI Design Management & Syst Engn Durgapur W Bengal India

In recent years, salient object detection (SOD) has achieved significant progress with the help of convolution neural network (CNN). Most of the state-of-the-art methods segment the salient object by either aggregating the multilevel features from the CNN module or introducing the refinement module along with the baseline network. However, these models suffer from simplicity bias, where neural networks converge to global minima using the simple feature and remain invariant to complex predictive features. Very few methods concentrate on the neurophysiological behaviour of visual attention. As per Gestalt psychology, humans tend to perceive the objects as a whole rather than focus on the discrete elements of that object. The law of Closure (closed contour) is one of the Gestalt axioms that states that if there is a discontinuity in the object's contour, we perceive the object as continuous in a smooth pattern. This paper proposes a two-way learning network, where Closure-guided Attention network (CGAN) and the Coarse Saliency networks (CSN) jointly supervise the feature-channel to mitigate the simplicity bias. Furthermore, a channel-wise attention residual network is incorporated in the Closure Guided module to alleviate the scale-space problem and generate smooth object contour. Finally, the closure map from CGAN fused with the coarse saliency map of the Coarse Saliency network generates a salient object. Experimental result on five benchmark datasets demonstrates the significant improvements in our approach over the state-of-the-art method.

关键词： Visual attention Saliency encoder-decoder network Gestalt Convolution neural network Multi-level fusion

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：