Microbial spatial distribution monitoring method and system based on multi-level microscopic scanning images
By combining multi-level microscopic scanning images and differential probability density functions, the problems of poor multi-fold adaptability and insufficient ecological specificity of models in microbial distribution monitoring are solved, and accurate monitoring and dynamic adaptation of microbial communities are achieved.
Patent Information
- Application Number
- CN202512035159.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-31
- Publication Date
- 2026-03-20
- Estimated Expiration
- 2045-12-31
AI Technical Summary
Existing methods for monitoring microbial distribution suffer from problems such as poor adaptability of multi-fold observation data, lack of ecological specificity in distribution models, insufficient quantification of fitting effects, and inability of parameters to dynamically adapt to community changes.
A multi-level microscopic scanning method is adopted to acquire images and their location information through multi-level grid traversal scanning. Differential probability density functions are designed in combination with microbial ecological types, and the model parameters are dynamically updated by incremental training using a rolling window.
It achieves precise capture and spatial distribution characterization of microorganisms of different sizes, improves model fitting accuracy and stability, and can dynamically adapt to changes in microbial communities.