Yemina

基于 YOLOv11 的酵母细胞自动计数桌面应用 YOLOv11-Powered Automated Yeast Cell Counting Desktop App

Yemina 是什么What is Yemina

Yemina 是一款开源的桌面应用，专为酵母细胞自动计数设计。它使用 YOLOv11 深度学习模型对显微图像中的酵母细胞进行像素级识别与计数，支持明场和相差成像。内置自动网格检测功能，可直接识别血球计数板的中方格区域。 Yemina is an open-source desktop application purpose-built for automated yeast cell counting. It leverages YOLOv11 deep learning for pixel-level cell detection in microscopy images, supporting both brightfield and phase contrast. Built-in automatic grid detection works directly with hemocytometer images.

核心功能Key Features

YOLOv11 深度学习识别YOLOv11 Deep Learning Detection

采用 YOLOv11（You Only Look Once v11）目标检测模型，在酵母细胞计数任务上实现像素级别的精准识别。相比传统图像处理方法（阈值分割、形态学运算），深度学习模型对光照不均、细胞重叠等复杂场景有更强的鲁棒性。 Uses YOLOv11, a state-of-the-art object detection architecture, for pixel-level yeast cell recognition. Compared to traditional image processing techniques like thresholding and morphological operations, the deep learning model handles uneven illumination, cell clumping, and complex backgrounds with greater robustness.

智能网格计数Automatic Grid Counting

自动识别血球计数板中的中方格区域，并遵循标准计数规则进行细胞计数。支持一键导出计数结果，适用于科研实验中的批量处理需求。 Automatically detects the center grid region on hemocytometer images and counts cells following standard hemocytometer counting rules. One-click export for batch processing in research workflows.

快速处理Fast Processing

多线程架构，10 秒内完成传统方法需要 10 分钟的处理工作。支持批量图像导入，适合高通量实验场景。 Multi-threaded architecture completes in seconds what takes minutes with traditional methods. Batch image import supported for high-throughput experiments.

技术细节Technical Details

与同类工具对比Comparison with Alternatives

vs 人工计数vs Manual Counting

人工计数耗时长、易疲劳、主观性强。Yemina 可在数秒内完成计数，结果可重复，消除主观差异。 Manual counting is time-consuming, fatigue-prone, and subjective. Yemina delivers consistent, reproducible results in seconds.

vs CellProfilervs CellProfiler

CellProfiler 是通用的图像分析平台，需要手动搭建 pipeline、调参。Yemina 专为酵母细胞计数优化，开箱即用，无需配置复杂的图像处理流程。 CellProfiler is a general-purpose image analysis platform requiring manual pipeline configuration and parameter tuning. Yemina is optimized specifically for yeast cell counting — works out of the box with no complex setup.

vs 传统图像处理vs Traditional Image Processing

阈值分割、边缘检测等传统方法对成像条件敏感，光照变化或细胞重叠时准确率显著下降。YOLOv11 模型经训练后可适应多种成像条件。 Traditional methods like thresholding and edge detection are sensitive to imaging conditions. YOLOv11 adapts to diverse microscopy conditions after training.

常见问题FAQ

Yemina 支持哪些显微成像方式？What microscopy methods does Yemina support?
支持明场和相差显微成像。模型经多种成像条件训练，适应性强。Supports brightfield and phase contrast microscopy. The model is trained on diverse imaging conditions.
Yemina 能处理多张图片的批量计数吗？Can Yemina batch process multiple images?
可以。支持批量导入图片并自动计数，结果一键导出。Yes. Batch import with automatic counting and one-click export.
Yemina 有论文发表吗？Is there a publication for Yemina?
相关论文正在审稿中（一作在审），核心方法 YOLOv11 模型和数据集已开源。Manuscript under review (first author). The core model and dataset are open-source on GitHub.
Yemina 是免费的吗？Is Yemina free?
完全免费开源。Completely free and open-source.

返回产品页Back to Product GitHub ↗

Yemina · 华中科技大学 Yemina · Huazhong University of Science and Technology