A5083953460- Advanced Image and Video Retrieval Techniques
- Sphingolipid Metabolism and Signaling
- Ion Transport and Channel Regulation
- Visual Attention and Saliency Detection
- Ion channel regulation and function
- Textile materials and evaluations
- Infrared Target Detection Methodologies
- Endoplasmic Reticulum Stress and Disease
- Video Surveillance and Tracking Methods
- Medical Image Segmentation Techniques
- Machine Fault Diagnosis Techniques
- Generative Adversarial Networks and Image Synthesis
- Gear and Bearing Dynamics Analysis
- Retinoids in leukemia and cellular processes
- Protein Degradation and Inhibitors
- Ubiquitin and proteasome pathways
- Advanced Neural Network Applications
- Image Enhancement Techniques
- Engineering Diagnostics and Reliability
- Genomics, phytochemicals, and oxidative stress
- Image Retrieval and Classification Techniques
- Video Analysis and Summarization
Dalian Polytechnic University
2025
Shenyang Aerospace University
2023
Ministry of Education of the People's Republic of China
2021
Anhui Medical University
2021
Abstract Remote sensing image target detection provides an effective and accurate data analysis tool for many application areas. Due to complex backgrounds, large differences in scales, missed of small targets, remote is challenging. In order enhance the model's understanding global information images, this paper proposes GFA module. This module can establish contextual connection images provide rich context help understand scene background which located, without being limited local...
Existing saliency object detection (SOD) methods struggle to satisfy fast inference and accurate results simultaneously in high resolution scenes. They are limited by the quality of public datasets efficient network modules for high-resolution images. To alleviate these issues, we propose construct a matting dataset HRSOM lightweight PSUNet. Considering mobile depolyment framework, design symmetric pixel shuffle module TRSU. Compared 13 SOD methods, proposed PSUNet has best objective...
Image editing approaches with diffusion models have been rapidly developed, yet their applicability are subject to requirements such as specific types (e.g., foreground or background object editing, style transfer), multiple conditions mask, sketch, caption), and time consuming fine-tuning of models. For alleviating these limitations realizing efficient real image we propose a novel technique that only requires an input target text for various including non-rigid edits without model. Our...
Aiming to precisely identify a compound fault of rolling bearing, the paper has contributed characteristic enhancement method by combing entropy weight (EWM) and intrinsic time scale decomposition (ITD). Firstly, effectively segregate frequency components in vibration signals, proper rotation (PRCs) were obtained decomposing signals based on ITD. Secondly, view fact that amplitude, variance correlation coefficient vary greatly bearing accompanied impact components, parameter evaluation...
Image matting is a fundamental technique in visual understanding and has become one of the most significant capabilities mobile phones. Despite development storage computing power, achieving diverse Artificial Intelligence Generated Content (AIGC) applications remains great challenge. To address this issue, we present an innovative demonstration automatic system called "Matting Moments" that enables image editing based on models different scenarios. Coupled with accurate refined subjects,...
Image editing approaches with diffusion models have been rapidly developed, yet their applicability are subject to requirements such as specific types (e.g., foreground or background object editing, style transfer), multiple conditions mask, sketch, caption), and time consuming fine-tuning of models. For alleviating these limitations realizing efficient real image we propose a novel technique that only requires an input target text for various including non-rigid edits without model. Our...
Existing saliency object detection (SOD) methods struggle to satisfy fast inference and accurate results simultaneously in high resolution scenes. They are limited by the quality of public datasets efficient network modules for high-resolution images. To alleviate these issues, we propose construct a matting dataset HRSOM lightweight PSUNet. Considering mobile depolyment framework, design symmetric pixel shuffle module TRSU. Compared 13 SOD methods, proposed PSUNet has best objective...