基于深度学习的金字塔变压器用于热压烧结陶糊的局部孔隙性分析
Zhongyi Xia1,2, Boqi Wu3, C Y Chan2
1College of Applied Technology, Shenzhen University, Shenzhen, Guangdong, China.
PloS one
|September 4, 2024
概括
我们开发了PSTNet,这是一种新的深度学习模型,用于在陶的扫描电子显微镜 (SEM) 图像中自动细分颗粒和毛孔. 与手动技术相比,这种方法显著提高了孔径测量精度.
科学领域:
- 材料科学 材料科学 材料科学
- 陶工程 陶工程
- 图像分析 图像分析
背景情况:
- 扫描电子显微镜 (SEM) 对于陶微观结构分析至关重要.
- 从SEM图像中手动提取多孔性是劳动密集型的,容易出错.
- 需要自动化方法来提高孔径量化的效率和准确性.
研究的目的:
- 开发和验证PSTNet (金字塔细分变压器网) 用于在陶SEM图像中自动化谷物和孔隙细分.
- 为了准确量化陶粒边界的多孔性.
- 提高陶微结构分析的效率和可靠性.
主要方法:
- 为了精确的细分,PSTNet合并了多尺度的特征地图.
- 该模型在Al2O3和Y2O3陶SEM数据集上进行了训练.
- 采用了一个联合损失函数,该函数包含了细分处罚交叉,平滑L1和SSIM损失.
- 改进的多头注意力机制用于解码器中的功能融合.
主要成果:
- 与基线方法相比,PSTNet在像素精度 (12.2%的增加) 和平均交叉点在整个欧盟 (mIoU) (25.5%的增加) 中取得了显著的改善.
- 该模型显示,Y2O3数据集的平均相对误差为6.9%,Al2O3数据集的平均误差为6.36%.
- 使用PSTNet进行的自动孔径测量显示出高精度和稳定性.
结论:
- 在陶SEM图像中,PSTNet提供了一种有效的自动化解决方案,用于在陶SEM图像中进行粒度和孔隙细分.
- 拟议的方法提高了陶毛孔度量化的准确性和效率.
- 这一进步有助于对陶材料特性进行更可靠的分析.
相关概念视频
01:18Porosity in Cement Paste
118
The porosity of concrete is a measure of the void spaces within its structure. These spaces impact its strength and durability significantly. When water and cement interact, a chemical reaction called hydration creates a semi-solid paste. This paste includes combined water, making up approximately 23% of the cement's dry mass, and gel water, which fills minuscule voids known as gel pores, accounting for about 28% of the cement gel volume.
The balance of water to cement in the mix is...
The balance of water to cement in the mix is...
118
01:23Pore Size Distribution
104
In concrete, the pore size distribution significantly influences the material's properties. Capillary pores, markedly larger than gel pores, form a vast network within partially hydrated cement paste, reducing the concrete's strength and increasing its permeability. This heightened permeability leads to a greater risk of damage from environmental factors like freeze-thaw cycles and chemical attacks, with the extent of vulnerability also being tied to the water-to-cement ratio.
Adequate...
Adequate...
104