Bright-Field Inspection for Bacteria Aggregates and Foreign Matter

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods struggle to accurately distinguish between isolated bacteria, aggregated bacteria, and foreign matters such as dust, dirt, and air bubbles in bacterial proliferation tests, leading to erroneous detection of bacterial proliferation.

Innovation Solution

An inspection device and method that utilizes image processing techniques, including binarization, contour extraction, and integrated signal difference analysis to differentiate between foreign matters and fine particle aggregates and accurately detect their presence and the degree of proliferation of fine particles.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of substance

If label-free detection method using bright-field microscope is used, then reagent costs are reduced, but foreign matters such as dust, dirt, and air bubbles are misdetected as bacteria

Engineering Contradiction:
Improvereagent costsVSAvoiddetection accuracy
Core Design Contradiction:
Loss of substanceVSMeasurement precision

Solution Approach 1:

The patent segments the detection process into multiple stages: initial image capture, foreign matter detection through binarization, contour extraction, and fine particle region detection. By dividing the detection workflow, the system can identify and exclude foreign matters while detecting bacteria, maintaining detection accuracy without using fluorescent reagents

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces image processing algorithms as an intermediary between the bright-field microscope and the detection result. The binarization and contour extraction processes act as mediators that differentiate between foreign matters and bacteria based on image characteristics, enabling accurate detection without fluorescent labeling

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If image processing is performed to detect bacterial proliferation, then detection sensitivity is improved, but it becomes difficult to distinguish between aggregated bacteria and foreign matters

Engineering Contradiction:
Improvedetection sensitivityVSAvoidobject differentiation information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent applies dynamic thresholding in the binarization process, adapting the threshold based on local image characteristics. This allows the system to maintain high sensitivity for detecting bacteria while dynamically adjusting to distinguish aggregated bacteria from foreign matters based on their different image patterns

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent applies different image processing operations to different regions of the image. By performing contour extraction and binarization with locally adapted parameters, the system can distinguish between aggregated bacteria and foreign matters in different image regions while maintaining overall detection sensitivity

Inventive Principle:
Principle #3Local quality

3Measurement precision

If binarization and contour extraction are performed to detect fine particles, then detection capability is improved, but foreign matters are still erroneously detected as bacteria

Engineering Contradiction:
Improvefine particle detection capabilityVSAvoiddetection reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent performs preliminary foreign matter detection through binarization before the main fine particle detection process. By identifying and excluding foreign matter regions in advance, the subsequent contour extraction and fine particle detection operate only on valid bacterial regions, improving both capability and reliability

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent extracts and removes foreign matter regions from the image through the binarization process before performing fine particle detection. By taking out the problematic foreign matter components, the system prevents erroneous detection while maintaining the ability to detect actual bacteria

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP4674942A1Inspection device for foreign matter and particle aggregates, and inspection method
Publication Date: 2026.01.07 HITACHI HIGH TECH CORP
  • EP4674942A1 patent drawingFigure 1A~1C
  • EP4674942A1 patent drawingFigure 2~3A
  • EP4674942A1 patent drawingFigure 3B~4

AI summary

Provided is an inspection device that can detect foreign matter and fine particle aggregates, and can accurately detect the degree of proliferation of fine particles. An inspection device 500 for aggregates comprises an image acquisition unit 11 that acquires an image of the inside of a container 511 holding a liquid containing fine particles, and detects the presence or absence of a fine particle aggregate. The inspection device 500 comprises: a foreign matter presence/absence determination unit 24 that binarizes an image, creates a first binarized image, and determines the presence or absence of foreign matter; a fine particle region detection unit 25 that extracts contours of the image to produce a contour-extracted image, binarizes a composite image of the extracted contour-extracted image and the image, creates a second binarized image, and detects a fine particle region; an integrated signal difference processing unit 16 that calculates the difference value between a first integrated signal, which was obtained by integrating the first binarized image, and a second integrated signal, which was obtained by integrating the second binarized image; and an aggregation presence/absence determination unit 17 that determines the presence or absence of aggregates by numerically analyzing the difference waveform of the difference value. The inspection device detects the presence or absence of at least two different types of objects contained in the image.