Category Ranking
98%
Total Visits
921
Avg Visit Duration
2 minutes
Citations
20
Article Abstract
Visual perception is one of the core technologies for achieving unmanned and intelligent mining in underground mines. However, the harsh environment unique to underground mines poses significant challenges to visible light-based visual perception methods. Multimodal fusion semantic segmentation offers a promising solution, but the lack of dedicated multimodal datasets for underground mines severely limits its application in this field. This work develops a multimodal semantic segmentation benchmark dataset for complex underground mine scenes (MUSeg) to address this issue. The dataset comprises 3,171 aligned RGB and depth image pairs collected from six typical mines across different regions of China. According to the requirements of mine perception tasks, we manually annotated 15 categories of semantic objects, with all labels verified by mining experts. The dataset has also been evaluated using classical multimodal semantic segmentation algorithms. The MUSeg dataset not only fills the gap in this field but also provides a critical foundation for research and application of multimodal perception algorithms in mining, contributing significantly to the advancement of intelligent mining.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12238567 | PMC |
| http://dx.doi.org/10.1038/s41597-025-05493-9 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Improving Generalized Visual Grounding with Instance-aware Joint Learning.
IEEE Trans Pattern Anal Mach Intell
September 2025
Generalized visual grounding tasks, including Generalized Referring Expression Comprehension (GREC) and Segmentation (GRES), extend the classical visual grounding paradigm by accommodating multi-target and non-target scenarios. Specifically, GREC focuses on accurately identifying all referential objects at the coarse bounding box level, while GRES aims for achieve fine-grained pixel-level perception. However, existing approaches typically treat these tasks independently, overlooking the benefits of jointly training GREC and GRES to ensure consistent multi-granularity predictions and streamline the overall process.
View Article and Find Full Text PDFLeveraging Multi-Text Joint Prompts in SAM for Robust Medical Image Segmentation.
IEEE J Biomed Health Inform
September 2025
The Segment Anything Model (SAM) has attracted considerable attention due to its impressive performance and demonstrates potential in medical image segmentation. Compared to SAM's native point and bounding box prompts, text prompts offer a simpler and more efficient alternative in the medical field, yet this approach remains relatively underexplored. In this paper, we propose a SAM-based framework that integrates a pre-trained vision-language model to generate referring prompts, with SAM handling the segmentation task.
View Article and Find Full Text PDFFine-grained multiclass nuclei segmentation with molecular empowered all-in-SAM model.
J Med Imaging (Bellingham)
September 2025
Vanderbilt University, Data Science Institute, Nashville, Tennessee, United States.
Purpose: Recent developments in computational pathology have been driven by advances in vision foundation models (VFMs), particularly the Segment Anything Model (SAM). This model facilitates nuclei segmentation through two primary methods: prompt-based zero-shot segmentation and the use of cell-specific SAM models for direct segmentation. These approaches enable effective segmentation across a range of nuclei and cells.
View Article and Find Full Text PDFSelfReg-UNet: Self-Regularized UNet for Medical Image Segmentation.
Med Image Comput Comput Assist Interv
October 2024
School of Computing and Augmented Intelligence, Arizona State University, AZ, USA.
Since its introduction, UNet has been leading a variety of medical image segmentation tasks. Although numerous follow-up studies have also been dedicated to improving the performance of standard UNet, few have conducted in-depth analyses of the underlying interest pattern of UNet in medical image segmentation. In this paper, we explore the patterns learned in a UNet and observe two important factors that potentially affect its performance: (i) irrelative feature learned caused by asymmetric supervision; (ii) feature redundancy in the feature map.
View Article and Find Full Text PDFArtificial Intelligence Automation of Echocardiographic Measurements.
J Am Coll Cardiol
August 2025
Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, California, USA; Department of Cardiology, Kaiser Permanente Santa Clara Medical Center, Santa Clara, California, USA. Electronic address:
Background: Accurate measurement of echocardiographic parameters is crucial for the diagnosis of cardiovascular disease and tracking of change over time; however, manual assessment requires time-consuming effort and can be imprecise. Artificial intelligence has the potential to reduce clinician burden by automating the time-intensive task of comprehensive measurement of echocardiographic parameters.
Objectives: The purpose of this study was to develop and validate open-sourced deep learning semantic segmentation models for the automated measurement of 18 anatomic and Doppler measurements in echocardiography.