Article
Computer Science, Artificial Intelligence
Bo Dong, Yan Zhou, Chuanfei Hu, Keren Fu, Geng Chen
Summary: The proposed Bidirectional Collaboration Network (BCNet) integrates effective multi-level feature fusion and multi-type feature aggregation into an edge-guided SOD framework. By utilizing Consistency Saliency Maximization (CSM) and Bounded Feature Fusion (BFF) modules, robust feature fusion and aggregation are achieved, leading to improved accuracy and speed compared to existing methods.
Article
Computer Science, Artificial Intelligence
Jin Zhang, Qiuwei Liang, Qianqian Guo, Jinyu Yang, Qing Zhang, Yanjiao Shi
Summary: The article introduces a Residual Refinement Network (R(2)Net) method for salient object detection, which improves the performance of salient object detection through the fusion strategy of multi-scale features and contextual features. Experimental results demonstrate that the proposed method performs excellently on multiple benchmark datasets.
IMAGE AND VISION COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Jin Zhang, Qiuwei Liang, Yanjiao Shi
Summary: This paper proposes a novel position prior attention network (PPANet) for salient object detection, achieving fast and accurate detection through the position prior attention module and context fusion module. Experimental results demonstrate that PPANet performs excellently in terms of accuracy and real-time performance.
IMAGE AND VISION COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Hongbo Bi, Ranwan Wu, Ziqi Liu, Huihui Zhu, Cong Zhang, Tian -Zhu Xiang
Summary: This paper proposes a cross-modal Hierarchical Interaction Network (HINet) to boost the salient object detection by excavating the cross-modal feature interaction and progressively multi-level feature fusion.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Cuili Yao, Lin Feng, Yuqiu Kong, Bo Jin, Yiwei Liu, Leheng Li
Summary: Recent deep learning-based methods in salient object detection have achieved impressive results by leveraging multi-level convolutional features, utilizing a Bi-directional CNN and DAI module to extract and integrate meaningful information, ultimately improving prediction accuracy and robustness.
Article
Computer Science, Artificial Intelligence
Cuili Yao, Lin Feng, Yuqiu Kong, Lin Xiao, Tao Chen
Summary: This paper presents a unified model combining CNN and Transformer for RGB and RGB-D salient object detection. The model utilizes Transformer to capture long-range relationships and CNN to extract multi-scale local features, achieving accurate saliency prediction. Experiments on RGB and RGB-D datasets demonstrate the superior performance of the proposed model.
Article
Computer Science, Artificial Intelligence
Jin Zhang, Yanjiao Shi, Qing Zhang, Liu Cui, Ying Chen, Yugen Yi
Summary: The article introduces a new neural network called ACFFNet for salient object detection, which includes MCA module, CFF module, and SR module, as well as a CCE loss to guide the network to focus on more detailed information. Experimental results show that the proposed method outperforms the state-of-the-art methods.
IMAGE AND VISION COMPUTING
(2022)
Article
Engineering, Multidisciplinary
Hongwei Wen, Kechen Song, Liming Huang, Han Wang, Junyi Wang, Yunhui Yan
Summary: Salient object detection using triple-modality information improves the detection effect. Locating and detecting the salient object accurately is crucial for subsequent grasping. We propose a method based on hierarchical two-stage modal fusion to address this problem.
Article
Computer Science, Artificial Intelligence
Caijuan Shi, Weiming Zhang, Changyu Duan, Houru Chen
Summary: This paper proposes a pooling-based feature pyramid network to enhance salient object detection performance. By designing U-shaped feature pyramid modules, pyramid pooling refinement modules, and channel attention modules, rich semantic information from high-level and low-level features can be captured effectively to improve detection accuracy.
IMAGE AND VISION COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Zun Li, Congyan Lang, Jun Hao Liew, Yidong Li, Qibin Hou, Jiashi Feng
Summary: The Feature Pyramid Network (FPN) based models have been effective in salient object detection, but often generate incomplete saliency maps due to indirect information propagation. The proposed Cross-layer Feature Pyramid Network (CFPN) improves progressive fusion in salient object detection by enabling direct cross-layer communication. Extensive experiments demonstrate that CFPN accurately locates salient regions and effectively segments object boundaries.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Bhagyashree Lad, Mohammad Farukh Hashmi, Avinash G. Keskar
Summary: This paper proposes a novel edge-directed salient object detection network that combines wavelet scattering network features with CNN-based features, enabling the network to capture both textural and high-level semantic information and improve the performance of salient object detection.
IMAGE AND VISION COMPUTING
(2023)
Article
Engineering, Electrical & Electronic
Jing Liu, Jiaxiang Wang, Weikang Wang, Yuting Su
Summary: This paper investigates the complementary roles of spatial and temporal information and proposes a novel dynamic spatiotemporal network for more effective fusion of spatiotemporal information. Experimental results demonstrate that the proposed method achieves superior performance in saliency detection.
DIGITAL SIGNAL PROCESSING
(2022)
Article
Computer Science, Artificial Intelligence
Qingping Zheng, Ling Zheng, Jiankang Deng, Ying Li, Changjing Shang, Qiang Shen
Summary: In this paper, a Transformer-based Hierarchical Dynamic Decoder (T-HDDNet) is proposed for salient object detection. The method utilizes a self-attention mechanism to extract features and has a powerful capability of learning global cues. With a dynamic dual upsampling mechanism and a dynamic feature fusion unit, it achieves accurate saliency maps of high resolution in a data-driven manner.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Hongfa Wen, Chenggang Yan, Xiaofei Zhou, Runmin Cong, Yaoqi Sun, Bolun Zheng, Jiyong Zhang, Yongjun Bao, Guiguang Ding
Summary: In this paper, a novel RGB-D saliency model called Dynamic Selective Network (DSNet) is proposed for salient object detection in RGB-D images, leveraging the complementary information between RGB images and depth maps. By optimizing multi-level and multi-scale information and refining boundaries, the proposed DSNet achieves competitive and excellent performance compared to existing state-of-the-art RGB-D SOD models.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Tianyou Chen, Jin Xiao, Xiaoguang Hu, Guofeng Zhang, Shaojie Wang
Summary: In this paper, a novel Adaptive Fusion Network (AFNet) is proposed to address the problem of RGB-D saliency detection. It designs three subnetworks to process RGB, depth, and fused features, and utilizes complementary cues to adaptively fuse the multi-modality features. Experimental results demonstrate that the proposed method outperforms 20 state-of-the-art counterparts on six commonly used benchmark datasets.
Article
Statistics & Probability
Ting Tian, Jingwen Zhang, Shiyun Lin, Yukang Jiang, Jianbin Tan, Zhongfei Li, Xueqin Wang
Summary: This study utilized interventions implemented in China and applied them to simulate the potential outcomes in South Korea, Italy, and the United States. The results showed that mild interventions in the early stage were effective in controlling the epidemic, while stricter measures were necessary for severe outbreaks.
JOURNAL OF APPLIED STATISTICS
(2023)
Article
Computer Science, Artificial Intelligence
Jiangxin Yang, Lingyu Wang, Lifei Ren, Yanpeng Cao, Yanlong Cao
Summary: This paper proposes an end-to-end light field angular super-resolution network by exploiting structure and scene information to mitigate the trade-off between angular and spatial resolution. It utilizes a Light Field ResBlock to extract epipolar plane image features and scene features, and employs 4D deconvolution to upsample the angular resolution based on these features. The refinement network further exploits LF structure and scene information to alleviate artifacts induced by too little angular information.
APPLIED INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Sizhe Xing, Guoqiang Li, Aolong Sun, Jiang Chen, Zhixue He, Junwen Zhang, Nan Chi
Summary: This study demonstrates a rate-flexible coherent TDM-PON in burst-mode based on PS-QAMmodulations and LO power adjustment technology, achieving a 300G peak rate and ultra-wide dynamic range. A new metric, DRNRP, is introduced to indicate the overall performance of achievable capacity and flexibility.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Guoqiang Li, Sizhe Xing, Junlian Jia, Zhongya Li, Junwen Zhang, Nan Chi
Summary: Coherent technology is a promising solution for future 100-Gb/s and even
200-Gb/s single-wavelength time-division multiplexing passive optical network (TDM-PON) systems. One of the key issues for coherent TDM-PON is the upstream burst-mode amplification.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Jianyang Shi, Zhongya Li, Junlian Jia, Ziwei Li, Chao Shen, Junwen Zhang, Nan Chi
Summary: In this paper, a learning-based waveform-to-waveform automatic equalization framework is proposed to overcome the shortcomings of small bandwidths and high harmonic interference levels in a seamless fiber-terahertz integrated communication system. Experimental results show that with this method, a data rate of 80.78 Gbps can be achieved for discrete multitone modulation signals over 5 km of fiber and 1 m of 209-GHz terahertz signals. Compared to an approach without preprocessing, a receiver sensitivity gain exceeding 1.3 dB is successfully achieved at a data rate of 60 Gbps. This proposed method is a promising scheme for future seamless fiber-terahertz integrated communication systems that require high speed and low cost.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Yanpeng Cao, Xi Tong, Fan Wang, Jiangxin Yang, Yanlong Cao, Sabin Tiberius Strat, Christel-Loic Tisse
Summary: This paper proposes a method for enhancing low-light visible images by utilizing complementary edge/texture features from thermal images. The proposed TGLLE-Net, which incorporates a CDC-MRF module and a TGCB module, effectively extracts and enhances features from both visible and thermal channels. Experimental results demonstrate that TGLLE-Net outperforms existing methods in terms of restoration accuracy, visual perception, and computational efficiency.
Article
Optics
Yanpeng Cao, Rui Liang, Wenbin Zhu, Bowen Zhao, Haotian Chen, Lingfeng Shen, Jiangxin Yang, Yanlong Cao, Jian Chen, Xin Li
Summary: This paper presents a complete framework for rendering and processing dynamic-excitation-based steady-state NLOS images. A virtual rendering pipeline is constructed to generate large-scale rendered images, and a physical hardware setup is built to acquire real-captured images. An end-to-end multi-branch CNN is proposed to reconstruct specular-reflected images of hidden objects. The effectiveness of the method is validated through qualitative and quantitative experiments.
OPTICS AND LASERS IN ENGINEERING
(2023)
Article
Chemistry, Analytical
Wenkai Yue, Ruixuan Liu, Peixian Li, Xiaowei Zhou, Yang Liu, Bo Yang, Yingxiao Liu, Xiaowei Wang
Summary: This paper investigates the effect of high-temperature annealing on the X-ray diffraction full width at half the maximum (XRD FWHM) of a 3.5 mu m-thick hydride vapor phase epitaxy-aluminum nitride (HVPE-AlN) (002) face. Tensile strain in the HVPE-AlN samples is gradually released with increasing annealing temperature. An aluminum oxynitride (AlON) region is generated at the contact interface between HVPE-AlN and sapphire when the annealing temperature exceeds 1700 degrees C, and the AlON structure conforms to the characteristics of Al5O6N.
Article
Geosciences, Multidisciplinary
Junwen Zhang, Yani Yan, Zhiqi Zhao, Congqiang Liu
Summary: This study analyzed the concentration and isotopic composition of lithium in river waters from the Niyang River, southern Tibetan Plateau. The results show that high isotopic values mainly come from geothermal water and silicate weathering. The study suggests that geothermal water input may mask the lithium isotope signal of silicate weathering in river water, affecting the accurate understanding of the relationship between lithium isotopic composition and weathering intensity in the river basin.
SCIENCE CHINA-EARTH SCIENCES
(2023)
Article
Environmental Studies
Shukui Tan, Bin Tong, Junwen Zhang
Summary: Land contracting is an important system in China and has made significant contributions to food security and agricultural development. However, the increase in land values has led to a rise in land contract disputes. This study used statistical methods, spatial analysis tools, and Markov Chains to analyze the temporal and spatial characteristics of land contract disputes in the Yangtze River Economic Belt. The results provide valuable insights for dealing with land contract disputes in relevant regions.
Proceedings Paper
Engineering, Electrical & Electronic
Boyu Dong, Junlian Jia, Guoqiang Li, Jianyang Shi, Haipeng Wang, Zhenzhou Tang, Junwen Zhang, Shilong Pan, Nan Chi
Summary: We propose and experimentally demonstrate a novel W-band photonic-based integration of sensing and communication system for the fiber-wireless integrated network, achieving adaptive sensing resolution and communication data-rates with flexible waveforms and TFDM resource allocation capability.
2023 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXHIBITION, OFC
(2023)
Proceedings Paper
Engineering, Electrical & Electronic
Guoqiang Li, An Yan, Sizhe Xing, Zhongya Li, Wangwei Shen, Jiaye Wang, Junwen Zhang, Nan Chi
Summary: To address burst-signal processing in downstream transmission when switching modulation formats, we propose and experimentally demonstrate a pilot-aided DSP scheme with continuous SOP tracking, carrier-phase recovery, and channel estimation in a 300G flexible CPON based on 4/16/64-QAMs.
2023 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXHIBITION, OFC
(2023)
Proceedings Paper
Engineering, Electrical & Electronic
Zhongya Li, Changle Huang, Junlian Jia, Guoqiang Li, Wangwei Shen, Jianyang Shi, Ziwei Li, Chao Shen, Junwen Zhang, Nan Chi
Summary: In this paper, we proposed and experimentally demonstrated a bit-wise end-to-end deep learning-based autoencoder for a fiber-THz integrated DFT-S-OFDM communication system at 209 GHz. Compared with the traditional DFT-S-OFDM system at 50 Gbps, it achieved a sensitivity gain of more than 5.5 dB.
2023 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXHIBITION, OFC
(2023)
Meeting Abstract
Gastroenterology & Hepatology
Zheng Zhang, Peng Li, Sheng Wang, Zhendong Jin, Yiqi Du, Aiming Yang, Yunlu Feng, Xiaoping Zou, Lei Wang, Xiaoyan Wang, Li Tian, Pinghong Zhou, Yiqun Zhang, Jun Liu, Zhen Ding, Junwen Zhang, Jian Yang, Siyu Sun, Shutian Zhang
JOURNAL OF GASTROENTEROLOGY AND HEPATOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Chupeng Yi, Yang Lu, Ziyue Zhao, Hengshuang Zhang, Bochao Zhao, Peixian Li, Xiaohua Ma, Yue Hao
Summary: This paper presents a design for a K-band high-efficiency power amplifier (PA) using modified resistive-reactive hybrid continuous modes (HCMs). The modified modes increase the real part of the fundamental impedance which helps reduce the impedance transformation ratio and make HCMs more practical. By combining this method with a compact output matching network (OMN), a high-performance and highly integrated PA can be achieved. Experimental results show that the designed PA has a saturation output power greater than 0.5 W and an average power-added efficiency (PAE) of 42%.
IEEE MICROWAVE AND WIRELESS TECHNOLOGY LETTERS
(2023)
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)