Effects of model architecture and learning strategies on deep learning-based recognition of activated sludge microscopic images and comparison with quantitative image analysis

📄 arXiv: 2609.08570v1 📥 PDF

作者: Suguru Hakoshima, Tomohiro Tobino, Fumiyuki Nakajima

分类: cs.CV

发布日期: 2026-09-08

💡 一句话要点

提出基于变换器的深度学习方法以提升活性污泥显微图像识别精度

🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 深度学习 显微图像分析 活性污泥 变换器模型 自监督学习 图像下采样 定量图像分析

📋 核心要点

  1. 现有的活性污泥显微图像分析方法多依赖于CNN和监督学习，缺乏对变换器模型的探索。
  2. 论文提出使用变换器架构和替代预训练方法，结合适当的下采样策略以提升分类精度。
  3. 研究结果显示，深度学习在活性污泥显微图像分类中表现优于传统的定量图像分析方法。

📝 摘要（中文）

显微图像分析被广泛认为是监测活性污泥的有效方法。近年来，基于深度学习的图像分析因其高性能而逐渐被采纳。然而，先前的研究很少探讨基于变换器的模型或自监督基础模型，且通常依赖于卷积神经网络（CNN）和监督的ImageNet预训练。此外，图像下采样的影响尚未得到充分研究。本研究旨在探讨模型架构和学习策略对活性污泥显微图像分析性能的影响，并定量比较深度学习与定量图像分析（QIA）的表现。结果表明，变换器架构和替代预训练方法在分类精度上表现出色，适当的下采样策略能有效提升识别准确性。最后，深度学习在准确性上超越了QIA。

🔬 方法详解

问题定义：本研究旨在解决活性污泥显微图像分析中现有方法的局限性，尤其是对变换器模型的忽视及下采样策略对性能的影响。

核心思路：通过引入变换器架构和自监督学习，结合对图像下采样的深入分析，提升显微图像分类的准确性。

技术框架：研究设计了一个包含数据准备、模型训练和性能评估的整体流程，使用三种不同类型的活性污泥样本进行分类。

关键创新：论文的主要创新在于首次系统性地比较了变换器模型与传统CNN在活性污泥图像分析中的表现，并提出了有效的下采样策略。

关键设计：在模型设计中，采用了变换器架构，结合不同的预训练方法，并在下采样时保持视野而非分辨率，以优化分类精度。

🖼️ 关键图片

fig_0
img_1
img_2

📊 实验亮点

实验结果显示，变换器架构在分类准确性上显著优于传统CNN，且深度学习方法的准确性超过了定量图像分析（QIA）。具体而言，适当的图像下采样策略能够有效提升识别精度，避免过度下采样导致的性能下降。

🎯 应用场景

该研究的成果可广泛应用于环境监测、污水处理等领域，帮助提升活性污泥的监测效率和准确性。未来，基于深度学习的显微图像分析方法有望在其他生物样本分析中发挥重要作用。

📄 摘要（原文）

Microscopic image analysis has long been recognized as a promising approach for monitoring activated sludge. In recent years, deep learning-based image analysis has been increasingly adopted in this field because of its high performance. However, previous studies on microscopic image analysis of activated sludge have rarely explored transformer-based models or self-supervised foundation models and have instead relied on CNNs and supervised ImageNet pretraining. In addition, previous studies often downsampled image sizes, but the effects of downsampling have not been sufficiently investigated, and the relationship between downsampling strategies and image analysis performance remains unclear. Furthermore, no study has quantitatively compared deep learning performance with quantitative image analysis (QIA), which was widely used before the emergence of deep learning. In this study, to examine how model architecture and learning strategies affect performance in microscopic image analysis of activated sludge and to quantitatively determine whether deep learning outperforms QIA, we prepared three types of activated sludge samples, classified their microscopic images, and evaluated classification accuracy. Our results showed that transformer-based architectures and alternative pretraining methods were effective in terms of classification accuracy. Our downsampling analysis showed that using overly small images reduced accuracy, but increasing image size beyond a certain point did not improve it further. In addition, the analysis indicated that, to achieve high classification accuracy, maintaining the field of view was a more effective downsampling strategy than maintaining resolution. Finally, our comparison between deep learning and QIA showed that deep learning outperformed QIA in terms of accuracy.