Article
Computer Science, Information Systems
Nannan Hu, Yue Ming, Chunxiao Fan, Fan Feng, Boyang Lyu
Summary: This research presents a Triple-Steam Feature Fusion Network (TSFNet) to improve the performance of image captioning by fusing visual representations from grid, region, and scene graph features. With a dual-level attention mechanism, the model explores both visual intrinsic properties and word-related attributes across different features. Experimental results demonstrate that the proposed model outperforms state-of-the-art image captioning approaches on the MSCOCO dataset, generating more accurate and abundant captions.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Software Engineering
Manoj Kumar Panda, Veerakumar Thangaraj, Badri Narayan Subudhi, Vinit Jakhetiya
Summary: This article introduces a unique and first attempt at visible and infrared image fusion using multi-scale decomposition and salient feature map detection. The proposed technique effectively handles uncertainty and retains maximum details of the sources at a multi-scale level. The combination of salient feature maps generates an image with complete information and reduced artifacts. The proposed algorithm outperforms existing techniques in terms of quantitative evaluation measures.
Article
Engineering, Electrical & Electronic
Yong Yang, Danjie Zhang, Weiguo Wan, Shuying Huang
Summary: A novel multiscale exposure image fusion method based on multivisual feature measurement and detail enhancement is proposed, which achieves better fusion performance by measuring the visual features of the source images and optimizing the weight maps.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Artificial Intelligence
Xuejian Li, Shiqiang Ma, Junhai Xu, Jijun Tang, Shengfeng He, Fei Guo
Summary: Automatic segmentation of medical images is crucial for disease diagnosis. This paper proposes a dual-path segmentation model called TranSiam for multi-modal medical images. The model utilizes parallel CNNs and a Transformer layer to extract features from different modalities, and aggregates the features using a locality-aware aggregation block.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Engineering, Electrical & Electronic
Jinyuan Liu, Xin Fan, Ji Jiang, Risheng Liu, Zhongxuan Luo
Summary: This article proposes a deep network for infrared and visible image fusion, incorporating a feature learning module and a fusion learning mechanism. It also introduces a new dataset for evaluation. Extensive experiments demonstrate the superiority of the proposed method in terms of various evaluation metrics.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Yu Ming, Zhang Jijun, Guo Yingchun, Zhang Meng, Wang Dan
Summary: A novel image aesthetic evaluation network based on multi-level attention fusion is proposed in this study, which extracts fine-grained features and adaptively fuses them according to the attention mechanism to protect aesthetic information and improve the visual perception of image retargeting. Experimental results show that the generated importance maps can preserve aesthetic information effectively and the resulting retargeting images have better visual perception compared to current methods.
LASER & OPTOELECTRONICS PROGRESS
(2021)
Article
Engineering, Mechanical
Zhenyu Han, Yue Zhuo, Yizhao Yan, Hongyu Jin, Hongya Fu
Summary: In this paper, a chatter detection method based on deep learning is proposed for the milling of thin-walled parts. By employing multi-channel signal features and attention mechanisms, the proposed method can accurately detect chatter during the milling process, which has been validated through experiments.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Chuanxu Wang, Huiru Wang
Summary: This paper introduces a self-attention mechanism for object detection, which effectively fuses the correlations among different features to improve the performance of object detection.
PATTERN RECOGNITION
(2023)
Article
Environmental Sciences
Wenhao Zhong, Jie Jiang, Yan Ma
Summary: The paper proposes a new method (L2AMF-Net) for accurate and robust lunar image patch matching. By using deep learning, the method overcomes difficulties such as illumination transformation, perspective transformation, resolution mismatch, and the lack of texture. L2AMF-Net achieves high matching accuracy and demonstrates excellent performance compared to other methods.
Article
Computer Science, Artificial Intelligence
Yucheng Shu, Jing Zhang, Bin Xiao, Weisheng Li
Summary: This study introduces a novel medical image segmentation framework, AFT-Net, which addresses existing issues by introducing an attention-based data fusion model and an Inception Res-Atrous Convolution block. Experiments demonstrate that this method is able to acquire image features with both diversity and quality, outperforming current state-of-the-art segmentation methods.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Yue Wu, Shutao Li
Summary: The paper introduces a novel multi-channel and single-channel fusion paradigm that can enhance the performance of multi-channel image denoising methods at low additional time-consuming cost. The principle of effectiveness lies in fusing the estimates of multi-channel and single-channel image denoisers in an underdetermined transform domain for better restoration of multi-channel images.
INFORMATION FUSION
(2022)
Article
Computer Science, Hardware & Architecture
Yuqin Song, Jitao Zhao, Chunliang Shang
Summary: This paper proposes an efficient multi-stage feature fusion defogging network based on the attention mechanism for effective defogging in complex environments. By integrating multiple attention techniques, a multi-branch defogging network is created to increase the effectiveness of the network model. Experimental results demonstrate that the proposed algorithm outperforms other sophisticated algorithms in terms of defogging performance and achieves better accuracy on the RESIDE and O-Haze datasets.
JOURNAL OF SUPERCOMPUTING
(2023)
Article
Chemistry, Analytical
Guoliang Yang, Hao Yang, Shuaiying Yu, Jixiang Wang, Ziling Nie
Summary: In this study, we propose a multi-scale dehazing network with dark channel priors (MSDN-DCP) to address the problems of incomplete dehazing, color deviation, and loss of detailed information. Experimental results show that the MSDN-DCP achieves superior dehazing compared to other algorithms in terms of objective metrics and visual perception.
Article
Engineering, Electrical & Electronic
Changcheng Wang, Dongming Zhou, Yongsheng Zang, Rencan Nie, Yanbu Guo
Summary: The article introduces a method of multi-focus image fusion using a deep learning structure, proposing a new Siamese multi-scale feature extraction module and an adaptive fusion strategy. The results show that the algorithm surpasses existing methods both quantitatively and qualitatively.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Software Engineering
YuJie Xu, YongJun Zhang, Zhi Li, ZhongWei Cui, YiTong Yang
Summary: In this paper, a novel multi-scale dehazing network via high-frequency feature fusion (HFMDN) is proposed to improve the haze removal performance. The network consists of a base network, a frequency branch network, a frequency attention module, and a refine block. Experimental results show that high-frequency information can significantly enhance dehazing performance, and the proposed method generates more natural and realistic haze-free images, especially in the contours and details of hazy images.
COMPUTERS & GRAPHICS-UK
(2022)
Article
Engineering, Industrial
W. S. Yip, S. To, HongTing Zhou
JOURNAL OF MANUFACTURING SYSTEMS
(2020)
Article
Computer Science, Artificial Intelligence
Wai Sze Yip, Suet To, Hongting Zhou
Summary: The article discusses the challenges of implementing sustainable practices in ultra-precision manufacturing (UPM) and suggests leveraging IoT technology to advance UPM towards sustainability.
JOURNAL OF INTELLIGENT MANUFACTURING
(2022)
Article
Engineering, Industrial
Hongting Zhou, Wai Sze Yip, Jingzheng Ren, Suet To
Summary: This paper utilizes the latent Dirichlet allocation (LDA) method to analyze the abstracts of relevant publications on sustainable ultra-precision machining (UPM), discovering four main research themes and exploring their contribution percentages. The findings suggest that the machining process is the most focal theme, while research on surface structure encompasses multiple topics. In addition, further investigation into the social aspect of sustainable UPM is needed.
JOURNAL OF MANUFACTURING SYSTEMS
(2022)
Review
Environmental Sciences
Wai Sze Yip, HongTing Zhou, Suet To
Summary: This study provides an overall overview of sustainable manufacturing research trends through thematic and bibliometric analysis, and identifies the critical periods and evolution of research works. It also demonstrates the research directions and advancements of sustainable manufacturing in 2020.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2022)
Article
Computer Science, Artificial Intelligence
Hongting Zhou, Wai Sze Yip, Jingzheng Ren, Suet To
Summary: This study proposed a new topic discovery model based on social network analysis and machine learning approach to discover undiscussed two-parameter relationships with high potential value in the field of sustainable ultra-precision machining. By analyzing the interactive relationships among parameters and applying a classification algorithm, the most valuable potential two-parameter topic in the area of sustainable UPM was found.
ADVANCED ENGINEERING INFORMATICS
(2022)
Article
Computer Science, Information Systems
Hongting Zhou, Wai Sze Yip, Jingzheng Ren, Suet To
Article
Engineering, Electrical & Electronic
Xueyu Han, Ishtiaq Rasool Khan, Susanto Rahardja
Summary: This paper proposes a clustering-based TMO method by embedding human visual system models to adapt to different HDR scenes. The method reduces computational complexity using a hierarchical scheme for clustering and enhances local contrast by superimposing details and controlling color saturation by limiting the adaptive saturation parameter. Experimental results show that the proposed method achieves improvements in generating high quality tone-mapped images compared to competing methods.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2024)
Article
Engineering, Electrical & Electronic
Zuopeng Zhao, Tianci Zheng, Kai Hao, Junjie Xu, Shuya Cui, Xiaofeng Liu, Guangming Zhao, Jie Zhou, Chen He
Summary: The research team developed a handheld phone detection network called YOLO-PAI, which successfully achieved real-time detection and underwent testing under various conditions. Experimental results show that YOLO-PAI reduces network structure parameters and computational costs while maintaining accuracy, outperforming other popular networks in terms of speed and accuracy.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2024)
Article
Engineering, Electrical & Electronic
Vivek Sharma, Ashish Kumar Tripathi, Purva Daga, M. Nidhi, Himanshu Mittal
Summary: In this study, a novel ClGan method is proposed for automated plant disease detection. The method reduces the number of parameters and addresses the issues of vanishing gradients, training instability, and non-convergence by using an encoder-decoder network. Additionally, an improved loss function is introduced to stabilize the learning process and optimize weights effectively. Furthermore, a new plant leaf classification method called ClGanNet is introduced, achieving 99.97% training accuracy and 99.04% testing accuracy using the least number of parameters.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2024)
Article
Engineering, Electrical & Electronic
Seongeun Kim, Chang-Ock Lee
Summary: This article introduces a method for segmenting individual teeth in human teeth images by using deep neural networks to obtain pseudo edge-regions and applying active contour models for segmentation.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2024)