Sample Tube Top Edge Enhancement for Accurate Robotic Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image-based detection methods in automated diagnostic analysis systems often erroneously detect stronger edge responses from objects other than the tops of sample tubes, such as barcode tags and tray components, leading to inaccurate handling and processing of samples.

Innovation Solution

The implementation of a convolutional neural network-based system that captures images of sample tube tops, intensifies the edges of the tubes while suppressing edges from other objects, and controls a robot for precise handling based on generated edge maps.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing image-based detection methods are used to detect sample tube tops, then the detection process is simple and fast, but the system erroneously detects stronger edge responses from other objects (barcode tags, tray components) instead of tube tops, leading to detection inaccuracies

Engineering Contradiction:

Improvedetection accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the detection approach by changing the parameter being detected from general edge strength to learned edge patterns through convolutional neural networks. The system learns to distinguish tube top edges from other objects by training on labeled data, fundamentally changing how edge detection parameters are interpreted and weighted.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces traditional mechanical/image processing-based edge detection algorithms with a learning-based convolutional neural network system. This substitution enables the system to automatically learn and adapt to distinguishing features of tube tops versus other objects, achieving higher precision without manual parameter tuning.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If traditional edge detection algorithms are applied to enhance tube top edges, then the processing is computationally efficient, but the system fails to suppress edge responses from other objects, resulting in false detections

Engineering Contradiction:

Improvedetection reliabilityVSAvoidcomputational energy

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent applies preliminary action by pre-training the convolutional neural network on large datasets of tube top images before deployment. This preliminary training phase enables the system to learn and encode discrimination patterns in advance, so that during actual detection, the system can reliably suppress false edges from other objects without requiring complex real-time computational adjustments.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11600058B2Methods and systems for learning-based image edge enhancement of sample tube top circles

Publication Date: 2023.03.07 SIEMENS HEALTHCARE DIAGNOSTICS INC

AI summary

Methods for image-based detection of the tops of sample tubes used in an automated diagnostic analysis system may be based on a convolutional neural network to pre-process images of the sample tube tops to intensify the tube top circle edges while suppressing the edge response from other objects that may appear in the image. Edge maps generated by the methods may be used for various image-based sample tube analyses, categorizations, and/or characterizations of the sample tubes to control a robot in relationship to the sample tubes. Image processing and control apparatus configured to carry out the methods are also described, as are other aspects.