检索结果-内蒙古大学图书馆

Medical image fusion method based on dense block and deep convolutional generative adversarial network

NEURAL COMPUTING & APPLICATIONS 2021年第12期33卷 6595-6610页

作者： Zhao, Cheng Wang, Tianfu Lei, Baiying Shenzhen Univ Natl Reg Key Technol Engn Lab Med Ultrasound Sch Biomed EngnHlth Sci Ctr Guangdong Key Lab Biomed Measurements & Ultrasoun Nanhai Ave 3688 Shenzhen 518060 Guangdong Peoples R China

Medical image fusion techniques can further improve the accuracy and time efficiency of clinical diagnosis by obtaining comprehensive salient features and detail information from medical images of different modalities. We propose a novel medical image fusion algorithm based on deep convolutional generative adversarial network and dense block models, which is used to generate fusion images with rich information. Specifically, this network architecture integrates two modules: an image generator module based on dense block and encoder-decoder and a discriminator module. In this paper, we use the encoder network to extract the image features, process the features using fusion rule based on the Lmax norm, and use it as the input of the decoder network to obtain the final fusion image. This method can overcome the weaknesses of the active layer measurement by manual design in the traditional methods and can process the information of the intermediate layer according to the dense blocks to avoid the loss of information. Besides, this paper uses detail loss and structural similarity loss to construct the loss function, which is used to improve the extraction ability of target information and edge detail information related to images. Experiments on the public clinical diagnostic medical image dataset show that the proposed algorithm not only has excellent detail preserve characteristics but also can suppress the artificial effects. The experiment results are better than other comparison methods in different types of evaluation.

关键词： Medical image fusion Deep convolutional GAN Dense block encoder-decoder Loss function

来源：评论

学校读者我要写书评

暂无评论

Lightweight semantic segmentation network with configurable context and small object attention

引用

FRONTIERS IN COMPUTATIONAL NEUROSCIENCE 2023年 17卷 1280640页

作者： Zhang, Chunyu Xu, Fang Wu, Chengdong Li, Jinzhao Northeastern Univ Fac Robot Sci & Engn Shenyang Peoples R China Shenyang Siasun Robot & Automat Co Ltd Shenyang Peoples R China Univ Chinese Acad Sci Changchun Inst Opt Fine Mech & Phys Beijing Peoples R China

The current semantic segmentation algorithms suffer from encoding feature distortion and small object feature loss. Context information exchange can effectively address the feature distortion problem, but it has the issue of fixed spatial range. Maintaining the input feature resolution can reduce the loss of small object information but would slow down the network's operation speed. To tackle these problems, we propose a lightweight semantic segmentation network with configurable context and small object attention (CCSONet). CCSONet includes a long-short distance configurable context feature enhancement module (LSCFEM) and a small object attention decoding module (SOADM). The LSCFEM differs from the regular context exchange module by configuring long and short-range relevant features for the current feature, providing a broader and more flexible spatial range. The SOADM enhances the features of small objects by establishing correlations among objects of the same category, avoiding the introduction of redundancy issues caused by high-resolution features. On the Cityscapes and Camvid datasets, our network achieves the accuracy of 76.9 mIoU and 73.1 mIoU, respectively, while maintaining speeds of 87 FPS and 138 FPS. It outperforms other lightweight semantic segmentation algorithms in terms of accuracy.

关键词： semantic segmentation lightweight network context feature enhancement small object attention encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Linear Contrast Enhancement Network for Low-Illumination Image Enhancement

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2023年 72卷 1页

作者： Zhou, Zhaorun Shi, Zhenghao Ren, Wenqi Xian Univ Technol Dept Comp Sci & Engn Xian 710048 Shaanxi Peoples R China Chinese Acad Sci Inst Informat Engn State Key Lab Informat Secur Beijing 100093 Peoples R China

Images captured under low-illumination conditions usually suffer from severe degradations, such as fading and low contrast, drastically affecting the performance of systems relying on images under low-illumination conditions. To address such problems, this study proposes a linear contrast enhancement network (LCENet) for low-illumination image enhancement. It consists of three subnets: two encoder-decoder-based subnets for gradient map restoration and brightness enhancement, respectively, and a backbone network for adaptive brightness and contrast adjustment. In addition, a linear contrast enhancement adaptive instance normalization (LCEAIN) module with linear contrast enhancement ability is proposed in the backbone network, which can avoid the problem of ignoring contrast enhancement when enhancing image brightness. Considerable evaluations on both synthetic and real low-illumination images show that the proposed method performs favorably against other existing similar methods. Moreover, our method can handle complex low-illuminance conditions and has good generalization for low-illuminance scenes with backlighting, night scenes with light sources, as well as underwater scenes with low illuminance. Code: https://***/zhouzhaorun/LCENet.

关键词： Brightness Image restoration Lighting Image enhancement Deep learning Image color analysis Task analysis Convolutional neural network encoder-decoder linear contrast enhancement low-illumination image enhancement

来源：评论

学校读者我要写书评

暂无评论

A self-supervised method of single-image depth estimation by feeding forward information using max-pooling layers

引用

VISUAL COMPUTER 2021年第4期37卷 815-829页

作者： Shi, Jinlong Sun, Yunhan Bai, Suqin Sun, Zhengxing Tian, Zhaohui Jiangsu Univ Sci & Technol Sch Comp Sci & Engn Zhenjiang Jiangsu Peoples R China Nanjing Univ State Key Lab Novel Software Technol Nanjing Peoples R China Ruizhi Informat Technol Co Ltd Zhenjiang Jiangsu Peoples R China

We propose an encoder-decoder CNN framework to predict depth from one single image in a self-supervised manner. To this aim, we design three kinds of encoder based on the recent advanced deep neural network and one kind of decoder which can generate multiscale predictions. Eight loss functions are designed based on the proposed encoder-decoder CNN framework to validate the performance. For training, we take rectified stereo image pairs as input of the CNN, which is trained by reconstructing image via learning multiscale disparity maps. For testing, the CNN can estimate the accurate depth information by inputting only one single image. We validate our framework on two public datasets in contrast to the state-of-the-art methods and our designed different variants, and the performance of different encoder-decoder architectures and loss functions is evaluated to obtain the best combination, which proves that our proposed method performs very well for single-image depth estimation without the supervision of ground truth.

关键词： encoder-decoder CNN Depth estimation Single image Self-supervision

来源：评论

学校读者我要写书评

暂无评论

FPA-Net: Frequency-Guided Position-Based Attention Network for Land Cover Image Segmentation

引用

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE 2023年第11期37卷 2354015-2354015页

作者： Rubel, Al Shahriar Shih, Frank Y. New Jersey Inst Technol Dept Informat Newark NJ 07102 USA New Jersey Inst Technol Dept Comp Sci Newark NJ 07102 USA Asia Univ Dept Comp Sci & Informat Engn Taichung Taiwan

Land cover segmentation has been a significant research area because of its multiple applications including the infrastructure development, forestry, agriculture, urban planning, and climate change research. In this paper, we propose a novel segmentation method, called Frequency-guided Position-based Attention Network (FPA-Net), for land cover image segmentation. Our method is based on encoder-decoder improved U-Net architecture with position-based attention mechanism and frequency-guided component. The position-based attention block is used to capture the spatial dependency among different feature maps and obtain the relationship among relevant patterns across the image. The frequency-guided component provides additional support with high-frequency features. Our model is simple and efficient in terms of time and space complexities. Experimental results on the Deep Globe, GID-15, and Land Cover AI datasets show that the proposed FPA-Net can achieve the best performance in both quantitative and qualitative measures as compared against other existing approaches.

关键词： Land cover segmentation encoder-decoder attention mechanism residual connection frequency-guided component U-Net

来源：评论

学校读者我要写书评

暂无评论

Explicit estimation of magnitude and phase spectra in parallel for high-quality speech enhancement☆

引用

NEURAL NETWORKS 2025年 189卷 107562页

作者： Lu, Ye-Xin Ai, Yang Ling, Zhen-Hua Univ Sci & Technol China Natl Engn Res Ctr Speech & Language Informat Proc Hefei Peoples R China

Phase information has a significant impact on speech perceptual quality and intelligibility. However, existing speech enhancement methods encounter limitations in explicit phase estimation due to the non-structural nature and wrapping characteristics of the phase, leading to a bottleneck in enhanced speech quality. To overcome the above issue, in this paper, we proposed MP-SENet, a novel Speech Enhancement Network that explicitly enhances Magnitude and Phase spectra in parallel. The proposed MP-SENet comprises a Transformer-embedded encoder-decoder architecture. The encoder aims to encode the input distorted magnitude and phase spectra into time-frequency representations, which are further fed into time-frequency Transformers for alternatively capturing time and frequency dependencies. The decoder comprises a magnitude mask decoder and a phase decoder, directly enhancing magnitude and wrapped phase spectra by incorporating a magnitude masking architecture and a phase parallel estimation architecture, respectively. Multi-level loss functions explicitly defined on the magnitude spectra, wrapped phase spectra, and short-time complex spectra are adopted to jointly train the MP-SENet model. A metric discriminator is further employed to compensate for the incomplete correlation between these losses and human auditory perception. Experimental results demonstrate that our proposed MP-SENet achieves state-of-the-art performance across multiple speech enhancement tasks, including speech denoising, dereverberation, and bandwidth extension. Compared to existing phase-aware speech enhancement methods, it further mitigates the compensation effect between the magnitude and phase by explicit phase estimation, elevating the perceptual quality of enhanced speech. Remarkably, for the speech denoising task, the proposed MP-SENet yields a PESQ of 3.60 on the VoiceBank+DEMAND dataset and 3.62 on the DNS challenge dataset.

关键词： Speech enhancement encoder-decoder Magnitude prediction Phase prediction Explicit estimation

来源：评论

学校读者我要写书评

暂无评论

Visual-guided attentive attributes embedding for zero-shot learning

引用

NEURAL NETWORKS 2021年 143卷 709-718页

作者： Zhang, Rui Zhu, Qi Xu, Xiangyu Zhang, Daoqiang Huang, Sheng-Jun Nanjing Univ Aeronaut & Astronaut Coll Comp Sci & Technol Nanjing 211106 Peoples R China

Zero-shot learning (ZSL) aims to learn a classifier for unseen classes by exploiting both training data from seen classes and external knowledge. In many visual tasks such as image classification, a set of high-level attributes that describe the semantic properties of classes are used as the external knowledge to bridge seen and unseen classes. While the attributes are usually treated equally by previous ZSL studies, we observe that the contribution of different attributes varies significantly over model training. To adaptively exploit the discriminative information embedded in different attributes, we propose a novel encoder-decoder framework with attention mechanism on the attribute level for zero-shot learning. Specifically, by mapping the visual features into a semantic space, the more discriminative attributes are emphasized with larger attention weights. Further, the attentive attributes and the class prototypes are simultaneously decoded to the visual space so that the hubness problem can be eased. Finally, the labels are predicted in the visual space. Extensive experiments on multiple benchmark datasets demonstrate that our proposed model achieves a significant boost over several state-of-the-art methods for ZSL task and comparative results for generalized ZSL task. (C) 2021 Elsevier Ltd. All rights reserved.

关键词： Zero-shot learning encoder-decoder Attributes Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Video-based driver action recognition via hybrid spatial-temporal deep learning framework

引用

MULTIMEDIA SYSTEMS 2021年第3期27卷 483-501页

作者： Hu, Yaocong Lu, Mingqi Xie, Chao Lu, Xiaobo Southeast Univ Sch Automat Nanjing 210096 Peoples R China Southeast Univ Minist Educ Key Lab Measurement & Control Complex Syst Engn Nanjing 210096 Peoples R China Nanjing Forestry Univ Coll Mech & Elect Engn Nanjing 210037 Peoples R China

Driver action recognition aims to distinguish normal driver action and some abnormal driver actions such as leaving the wheel, talking on the phone, diving with smoking, etc. For the purpose of traffic safety, studies on the computer vision technologies for driver action recognition have become especially meaningful. However, this issue is far from being solved, mainly due to the subtle variations between different driver action classes. In this paper, we present a new video-based driver action recognition approach based on the hybrid spatial-temporal deep learning framework. Specifically, we first design an encoder-decoder spatial-temporal convolutional neural network (EDSTCNN) to capture short-term spatial-temporal representation of driver actions jointly with optical flow prediction. Second, we exploit the feature refinement network (FRN) to refine the short-term driver action feature. Then, convolutional long short-term memory network (ConvLSTM) is employed for long-term spatial-temporal fusion. Finally, the fully connected neural network (FCNN) is used for final driver action recognition. In our experiment, we validate the performance of the proposed framework on our self-created datasets, including a simulated driving dataset and a real driving dataset. Extensive experimental results illustrate that the proposed hybrid spatial-temporal deep learning framework obtains the highest accuracy in multiple driver action recognition datasets (98.9% on SEU-DAR-V1 dataset and 97.0% on SEU-DAR-V2 dataset).

关键词： Driver action encoder-decoder Spatial-temporal Attention Convolutional long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Sketch2Photo: Synthesizing photo-realistic images from sketches via global contexts

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2023年第PartA期117卷

作者： Liu, Heng Xu, Yao Chen, Feng Anhui Univ Technol Maanshan 243032 Peoples R China Anhui Engn Res Ctr Intelligent Applicat & Secur In Maanshan 243032 Peoples R China

Sketch-to-image synthesis aims to generate realistic images that match the input sketches or edge maps exactly. Most known sketch-to-image synthesis methods use various generative adversarial networks (GANs) that are trained with numerous pairs of sketches and real images. Because of the convolution locality, the low-level layers of the generators in these GANs lack global perception ability, causing feature maps derived from them easily to overlook global cues. Since the global receptive field is crucial for acquiring the non-local structures and features of sketches, the absence of global contexts will impact the generation of high-quality images. Some recent models turn to self-attention to construct global dependencies. However, they are not viable for large feature maps for the quadratic computational complexity concerning the size of feature maps. To address these problems, in this work, we propose Sketch2Photo - a new image synthesis approach that can capture global contexts as well as local features to generate photo-realistic images from weak or partial sketches or edge maps. We employ fast Fourier convolution (FFC) residual blocks to create global receptive fields in the bottom layers of the network and incorporate Swin Transformer block (STB) units to obtain long-range global contexts for large-size feature maps efficiently. We also present an improved spatial attention pooling (ISAP) module to relax the strict alignment requirements between incomplete sketches and generated images. Quantitative and qualitative experiments on multiple public datasets demonstrate the superiority of the proposed approach over many other sketch-to-image synthesis methods. The project code is available at https://***/hengliusky/Skecth2Photo.

关键词： Sketch-based image synthesis Fast Fourier convolution Swin transformer encoder-decoder Global contexts

来源：评论

学校读者我要写书评

暂无评论

Improved Short-term Dense Bottleneck network for efficient scene analysis

引用

COMPUTER VISION AND IMAGE UNDERSTANDING 2023年第1期235卷

作者： Singha, Tanmay Pham, Duc-Son Krishna, Aneesh Curtin Univ Perth WA 6102 Australia

Visual scene understanding mainly depends on pixel-wise classification obtained from a deep convolutional neural network. However, existing semantic segmentation models often face difficulties in real-time appli-cations due to their large network architecture. Although there are real-time semantic segmentation models available, their shallow backbone can degrade the performance considerably. This paper introduces SDBNetV2, a lightweight semantic segmentation model designed to improve real-time performance without increasing computational costs. A key contribution is a novel Short-term Dense Bottleneck (SDB) module in the encoder, which provides varied field-of-views to capture different geometrical objects in a complex scene. Additionally, we propose dense feature refinement and improved semantic aggregation modules at the decoder end to enhance contextualization and object localization. We evaluate the proposed model's performance on several indoor and outdoor datasets in structured and unstructured environments. The results show that SDBNetV2 achieves superior segmentation performance over other real-time models with less than 2 million parameters.

关键词： Semantic segmentation Convolutional neural network Scene understanding Real-time Machine learning encoder-decoder

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：