Chromatography Autosampler Cell Identification via Machine Vision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autosamplers require manual labeling and placement of vials, leading to high error rates due to the complexity of managing multiple vials in a sample tray, with little to no chain-of-custody assurance once vials are loaded.
Innovation Solution
A method and system utilizing a machine vision module within the autosampler to scan the end surfaces and side walls of sample cells, combined with a cell gripper assembly to grip, lift, and rotate the cells, allowing for automated identification and positioning of sample cells within the autosampler.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual labeling and placement of vials is used, then users can identify samples, but error rates increase due to complexity of managing multiple vials
Solution Approach 1:
The patent replaces the manual mechanical process of labeling and tracking vials with an automated machine vision system using cameras and image processing algorithms. The system automatically captures images of vial labels, processes them through OCR or pattern recognition, and matches them to the correct positions in the sample tray, eliminating manual intervention and reducing errors.
Solution Approach 2:
The system enables self-service by allowing the autosampler to automatically identify and locate samples without user intervention. The machine vision system independently performs label recognition, position determination, and verification, making the system self-sufficient in sample identification tasks.
2Reliability
If manual placement of vials in sample tray is used, then users can load samples, but chain-of-custody assurance is lost once vials are placed
Solution Approach 1:
The patent implements feedback by continuously monitoring and verifying sample positions through machine vision. The system captures images of the sample tray, automatically identifies vial positions and labels, and provides real-time verification to the control system. This closed-loop feedback ensures chain-of-custody tracking is maintained throughout the autosampler operation.
Solution Approach 2:
The machine vision system acts as an intermediary between the physical sample tray and the digital control system. It bridges the gap by capturing visual information from the physical world, processing it through image recognition, and translating it into digital data that the control system can use for tracking and verification.
3Measurement precision
If users manually track vial positions, then sample locations can be recorded, but significant user error occurs in placement
Solution Approach 1:
The patent replaces manual visual tracking and recording with an automated machine vision system. Cameras capture high-resolution images of the sample tray, and image processing algorithms automatically determine the precise position of each vial. This eliminates the need for users to manually track and record positions, thereby eliminating human error in position measurement.
4Productivity
If sample trays have capacities of 50 or more vials, then throughput increases, but error probability increases due to complexity
Solution Approach 1:
The system enables self-service by allowing the autosampler to automatically identify, verify, and track each vial independently. The machine vision system autonomously processes all vials in the tray, recognizing labels and determining positions without user intervention, which is essential for maintaining reliability when handling large numbers of samples.
Solution Approach 2:
The patent implements comprehensive feedback by monitoring all vial positions and labels simultaneously through the machine vision system. The system provides real-time verification for each vial placement, comparing actual positions against the expected configuration, and can alert or correct errors automatically, ensuring high reliability even with 50 or more vials in the tray.
Data Source
Figure 1
Figure 2A~2F
Figure 3A~4
AI summary
Methods and systems are provided for the identification of sample cells in a sample tray that is placed in a chromatography autosampler. A cell gripper and labels are also provided to facilitate such identification.