Related Experiment Video
Updated: May 6, 2026
12:08
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
Published on: August 13, 2014
24.5K
Developing WasteSAM: A novel approach for accurate construction waste image segmentation to facilitate efficient
1Department of Architectural Engineering, School of Architecture, Dankook University, Yongin-si, Gyeonggi-do, South Korea.
Summary
WasteSAM, an improved segment anything model (SAM), enhances construction waste image segmentation for better recycling. It significantly outperforms the original SAM, improving waste classification accuracy.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Environmental Engineering
Background:
- Growing construction waste necessitates advanced sorting and recycling solutions.
- Automated segmentation models are crucial for efficient waste management.
Purpose of the Study:
- To introduce WasteSAM, an enhanced segmentation model for construction waste.
- To improve the accuracy and efficiency of construction waste classification and recycling.
Main Methods:
- Developed WasteSAM, an iteration of the segment anything model (SAM).
- Utilized a dataset of over 15,000 masks for five construction material categories.
- Integrated stereo-image techniques to enhance 3D structural discernment of waste materials.
Main Results:
- WasteSAM demonstrated superior segmentation performance compared to the original SAM.
- Achieved an average improvement of 23.9% in dice similarity coefficient.
- Showed a 30.0% improvement in normalized surface distance metrics.
- Accurately handled complex textures, diverse lighting, and object interactions.
Conclusions:
- WasteSAM offers significantly improved segmentation for construction waste.
- Stereo-image integration enhances 3D understanding and classification precision.
- High-quality, diverse datasets reflecting real-world conditions are vital for model development.
Related Concept Videos
01:29Aggregates Classification
1.0K
Aggregate classification is generally based on its size, petrographic characteristics, weight, and source. Size classification ranges from coarse to fine aggregates, defined by the size of the particles. Coarse aggregates are particles that do not pass through ASTM sieve No. 4, and aggregates that pass through the sieve are fine aggregates.
Petrographic classification groups aggregates based on common mineralogical characteristics. Some of the common mineral groups found in aggregates are...
Petrographic classification groups aggregates based on common mineralogical characteristics. Some of the common mineral groups found in aggregates are...
1.0K
01:28Microbial Bioremediation of Plastics
142
Polyethylene terephthalate (PET) is a synthetic polymer widely utilized in the packaging industry, particularly for bottles and containers. Due to its chemical stability and durability, PET accumulates in the environment, contributing significantly to plastic pollution. It comprises repeating units of terephthalic acid and ethylene glycol, resulting in a semi-crystalline structure that is resistant to natural degradation processes.A notable breakthrough in plastic biodegradation came with the...
142