一种有效的微观图像增强方法
Wanying Li1,2, Linhe Yang3, Guobei Peng4
1Guangxi Colleges and Universities Key Laboratory of Intelligent Software, Wuzhou University, Wuzhou, 543002, China.
Scientific reports
|March 26, 2025
概括
这项研究引入了一种新的微观图像增强方法,用于少数拍摄学习 (MIAA-FSL),以解决中药草 (CMH) 识别中的小样本大小. 该方法显著提高了识别罕见特征的准确性,提高了整体识别性能.
科学领域:
- 显微镜学和计算生物学
- 植物学中的人工智能
- 药物鉴定和机器学习
背景情况:
- 鉴定中国药草 (CMH) 面临挑战,因为大量的物种和难以收集显微镜图像,导致小样本大小.
- 某些细胞特征的稀缺性 (低至0.5%) 阻碍了深度学习和少量学习模型的有效性.
- 扩大罕见特征的数据对于提高CMH识别准确度至关重要.
研究的目的:
- 为少数拍摄学习 (MIAA-FSL) 提出一种有效的微观图像增强方法,以解决CMH识别中的数据稀缺性和类失衡问题.
- 开发一种有条件引导的微观图像生成模型 (CGMIGM),用于生成罕见的特征.
- 整合数据增强 (SSLDAM) 的半监督学习,以提高受损或模糊的微观图像的可用性.
主要方法:
- 使用无噪声扩散概率模型 (DDPM) 生成罕见特征并减轻类失衡的条件指导.
- 半监督学习和伪标签生成以增强和利用受损,模糊或难以辨别的显微镜图像.
- 开发了微观图像增强为少数镜头学习 (MIAA-FSL) 的方法,结合了CGMIGM和SSLDAM.
主要成果:
- 与MIR+DDPM方法相比,MIAA-FSL方法在识别准确度上平均提高了24%.
- 识别罕见特征的准确性显著提高,从45.5%增加到87.0%.
- 有效地减轻了在CMH显微镜图像分析中使用少数样本对象检测的挑战.
结论:
- 拟议的MIAA-FSL方法有效地解决了CMH微观图像识别中小样本大小和类不平衡的问题.
- 条件生成模型和半监督学习的组合增强了罕见特征的识别,并改善了模型的整体性能.
- 这种方法为准确的CMH识别提供了可行的解决方案,特别是在数据有限和特征稀缺的场景中.
相关概念视频
Overview of Microscopy Techniques
9.6K
The early pioneers of microscopy opened a window into the invisible world of microorganisms. In 1830, Joseph Jackson Lister created an essentially modern light microscope. The 20th century saw the development of microscopes that leveraged nonvisible light, such as fluorescence microscopy that uses an ultraviolet light source and electron microscopy that uses short-wavelength electron beams. These advances significantly improved magnification, image resolution, and contrast. By comparison, the...
9.6K
Super-resolution Fluorescence Microscopy
6.8K
Super-resolution fluorescence microscopy (SRFM) provides a better resolution than conventional fluorescence microscopy by reducing the point spread function (PSF). PSF is the light intensity distribution from a point that causes it to appear blurred. Due to PSF, each fluorescing point appears bigger than its actual size, and it is the PSF interference of nearby fluorophores that causes the blurred image. Various approaches to achieving higher resolution through SRFM have recently been...
6.8K
Imaging Biological Samples with Optical Microscopy
4.5K
Optical microscopy uses optic principles to provide detailed images of samples. Antonie van Leeuwenhoek designed the first compound optical microscope in the 17th century to visualize blood cells, bacteria, and yeast cells. In 1830, Joseph Jackson Lister created an essentially modern light microscope. The 20th century saw the development of microscopes with enhanced magnification and resolution.
In optical microscopy, the specimen to be viewed is placed on a glass slide and clipped on the stage...
In optical microscopy, the specimen to be viewed is placed on a glass slide and clipped on the stage...
4.5K


