Fast detection of micro-objects using scanning electrochemical microscopy based on visual recognition and machine...

Area of Science:

• Electrochemistry
• Surface Science
• Microscopy

Background:

• Scanning electrochemical microscopy (SECM) is a powerful technique for characterizing surface electrochemical properties using an ultramicroelectrode (UME).
• Current SECM methods face limitations including slow imaging speeds and significant user intervention, hindering full utilization of its potential.

Purpose of the Study:

• To develop an automated SECM system incorporating visual recognition and machine learning.
• To enhance imaging speed and reduce the need for constant user presence during analysis.
• To accurately detect micro-objects and determine their electrochemical activity.

Main Methods:

• Implementation of visual recognition and machine learning algorithms for micro-object detection and electrochemical activity assessment.
• Development of an image reconstruction method utilizing multiple approach curves for faster scanning.
• Integration of commercially available modules, cost-effective components, and established software solutions.
• Creation of an original control and data fusion algorithm for system operation.

Main Results:

• Achieved faster scanning speeds through image reconstruction from approach curves.
• Successfully detected micro-objects and identified their active electrochemical areas.
• Demonstrated reduced scanning time and diminished user presence requirements.
• Developed a functional automated SECM system with integrated visual recognition.

Conclusions:

• The developed automated SECM system effectively overcomes the limitations of traditional methods.
• Visual recognition and machine learning significantly improve the efficiency and accessibility of SECM analysis.
• This automated approach broadens the applicability of SECM for surface characterization.