An active sludge microorganism image deblurring enhancement method based on degenerate competition

By constructing a degradation competition kernel library and an adaptive fuzzy kernel model, the problem of inconsistent degradation types in different regions of activated sludge microscopic images was solved, improving image clarity and structural continuity, and making it suitable for microbial image analysis in wastewater treatment.

CN122243814APending Publication Date: 2026-06-19UNIV OF JINAN
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
UNIV OF JINAN
Filing Date
2026-05-22
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing microscopic image enhancement methods are ill-suited to addressing the inconsistency in degradation types across different regions of activated sludge microscopic images. This results in difficulty in clearly identifying microbial outlines, cilia, flagella, stipes, nematode edges, or filamentous fungal structures, impacting the accuracy of subsequent observation and automated identification analysis.

Method used

By constructing a confidence map of microbial structure, a background interference map of flocculents, and a response map of slender structures, a candidate degenerate kernel library is established for microscopic out-of-focus blur kernels, liquid disturbance motion blur kernels, and random jitter blur kernels. Degenerate competition weights are generated based on local recovery residuals to generate spatial adaptive blur kernels. A variational deblurring energy model is constructed to achieve adaptive deblurring enhancement.

Benefits of technology

It improves the clarity, structural continuity, and recognizability of activated sludge microbial images, reduces the impact of complex floc backgrounds on recognition results, and is suitable for observing the operational status of wastewater treatment and identifying microbial morphology.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122243814A_ABST
    Figure CN122243814A_ABST
Patent Text Reader

Abstract

This invention provides a deblurring and enhancement method for activated sludge microbial images based on degradation competition, belonging to the field of image enhancement. The method aims to construct a microbial structure confidence map, a floc background interference map, and a slender structure response map by performing grayscale normalization, noise screening, and field-of-view brightness correction on activated sludge microscopic images. A candidate degradation kernel library is established, consisting of a microscopic out-of-focus blur kernel, a liquid disturbance motion blur kernel, and a random jitter blur kernel. Degradation competition weights are generated based on local recovery residuals and fused to obtain a spatially adaptive blur kernel. A variational deblurring energy model is constructed and solved alternately by combining adaptive diffusion parameters and spatially adaptive p-values ​​to improve the clarity, structural continuity, and recognizability of the microbial images.
Need to check novelty before this filing date? Find Prior Art