AngstromPro: A versatile software for massive N-dimensional STM data management, visualization and in-depth analysis

This paper introduces AngstromPro, a versatile, open-source Python software designed to streamline the management, visualization, and analysis of massive N-dimensional Scanning Tunneling Microscope (STM) datasets through a modular architecture that enhances reproducibility, efficiency, and extensibility.

The Gist

Imagine you are a detective trying to solve a mystery, but instead of fingerprints, your clues are atoms. You have a super-powerful microscope called a Scanning Tunneling Microscope (STM) that can see individual atoms. However, this microscope is so powerful that it doesn't just take one picture; it takes millions of pictures at once, creating a massive, multi-layered 3D data mountain.

The problem? Scientists were trying to climb this mountain with a rusty spoon. They had to use five or six different software programs just to clean up the data, fix the blurry parts, and measure the atoms. It was messy, slow, and prone to mistakes.

Enter AngstromPro. Think of it as the "Swiss Army Knife" or a "Super-Organized Kitchen" for these atomic detectives.

Here is how AngstromPro works, explained simply:

1. The "Master Organizer" (The Kitchen)

Imagine a kitchen where every ingredient has its own labeled jar, and every tool has a specific spot. Before AngstromPro, scientists were throwing ingredients on the floor and using a different knife for every task.

• What it does: AngstromPro is a single software that manages all your data. It keeps a "Global List" of all your files and a "Local List" for the specific project you are working on right now.
• The Benefit: It keeps your workspace tidy. You never lose track of which version of your data you are looking at, and it remembers every single step you took (like a cooking recipe that writes itself down as you go).

2. The "Two-Screen Detective" (The Dual-Panel)

This is the software's coolest trick. Imagine you are looking at a blurry photo of a crime scene (the Main Panel). Usually, you'd have to open a second program to see the "fingerprint analysis" (the math behind the photo).

• What it does: AngstromPro has a Main Screen and an Auxiliary Screen side-by-side. When you look at the blurry photo on the left, the right screen automatically shows you the "math version" (the Fourier Transform) of that same photo.
• The Analogy: It's like having a detective who can look at a blurry face and instantly see a high-definition sketch of the same person next to it. You can zoom in on the blurry photo, and the sketch zooms in with you. This makes finding patterns (like atomic crystals) incredibly fast.

3. The "Magic Fix-It" Tools

The software comes with built-in tools to fix common problems, like a photo editor that knows exactly how to fix a crooked horizon.

• Background Subtract: Sometimes the microscope is tilted, making the whole image look like a ramp. AngstromPro flattens the ramp so you can see the atoms clearly.
• Perfect Lattice: Atoms sometimes look slightly squished or twisted because the machine moved a tiny bit. AngstromPro can "stretch" the image back into a perfect honeycomb shape, just like straightening a crooked picture on a wall.
• Lawler-Fujita Correction: This is the "super-fix." It corrects tiny, invisible wobbles in the data caused by temperature changes or the machine breathing. It turns a fuzzy, blurry image into a crystal-clear, sub-atomic sharp image.

4. The "Lego Block" System (Customization)

This is the most important part for the future. AngstromPro isn't a locked box; it's a Lego set.

• For Regular Users: You can just click buttons to do the standard fixes.
• For Experts: If you are a scientist who invented a new way to analyze data, you can write your own "Lego block" (a custom script) and snap it right into the software. You don't need to rebuild the whole kitchen; you just add your new tool to the drawer.
• Open Source: The code is free for everyone to see and improve. It's like a community garden where everyone can plant new flowers (algorithms) for everyone else to enjoy.

Why Does This Matter?

Before AngstromPro, analyzing a single experiment could take days of switching between programs and copying data back and forth. It was like trying to build a house by running between the lumber yard, the hardware store, and the paint shop.

AngstromPro brings the lumber, the tools, and the paint into one garage. It makes the process faster, more accurate, and reproducible. This means scientists can spend less time wrestling with software and more time discovering new quantum materials, superconductors, and the secrets of the universe.

In short: AngstromPro is the ultimate digital assistant that turns a chaotic mountain of atomic data into a clear, organized, and beautiful map of the microscopic world.

Technical Summary

1. Problem Statement

Modern Scanning Tunneling Microscopes (STMs) have evolved to operate under extreme conditions (ultra-low temperatures, high magnetic fields) and utilize advanced techniques (lock-in spectroscopy, spin resonance) to generate massive N-dimensional datasets. These datasets vary by spatial coordinates, electron energy, temperature, and magnetic field.

Despite hardware advancements, software development has lagged, creating several critical bottlenecks:

• Fragmented Workflows: Researchers often rely on a patchwork of tools (e.g., WSxM, Gwyddion, custom Python scripts) to manage, visualize, and analyze data, leading to disjointed processes.
• Lack of Reproducibility: Many tools lack robust processing history tracking, making it difficult to reproduce complex analysis steps.
• Limited Extensibility: Existing software often lacks open-source, modular architectures, making it difficult for users to integrate custom algorithms or adapt to new experimental setups.
• Inefficiency in N-Dimensional Handling: Current tools struggle to efficiently visualize and process high-dimensional data (e.g., $I(V)$ maps, 3D topographies) and perform complex corrections like lattice distortion or drift correction.

2. Methodology and Software Architecture

AngstromPro is an open-source, Python-based software built on NumPy, Matplotlib, and PyQt, developed within the Spyder IDE (Anaconda distribution). Its design philosophy emphasizes high cohesion and low coupling, separating the Graphical User Interface (GUI) from data processing algorithms.

Core Architectural Features:

• Modular & Extensible Design:
  • Top-Level Manager: A central "Modules & Variables Manager" oversees a Global Variables List (containing Unified Data Structure or UDS variables) and a list of active sub-modules.
  • UDS (Unified Data Structure): A standardized data container holding the N-dimensional data array, metadata (info), processing history, and variable names. This ensures data integrity and traceability.
  • Sub-Modules: Functional modules (e.g., Image2U3, Data Browser) inherit from a base class, maintaining their own Local Variables List. This prevents workspace clutter and allows multiple instances of the same module to run simultaneously.
• Dual-Panel GUI (Main–Auxiliary Collaboration):
  • The flagship Multiple 2D Images Visualizer & Analyzer (Image2U3) features a dual-panel layout.
  • MAIN Panel: Displays the primary dataset.
  • AUXILIARY Panel: Automatically displays the Fast Fourier Transform (FFT) of the MAIN panel data (or a second dataset).
  • Synchronization: The panels support synchronized cursors, zooming, layer switching, and point marking, facilitating binary operations (e.g., image subtraction, registration) and feature selection (Bragg peaks).
• Data Browser: An interactive interface for browsing local datasets in various formats (Nanonis .sxm, .3ds, MATLAB .mat, etc.). It uses a snapshot caching mechanism to handle large datasets efficiently without loading full data into memory until selected.
• User Extensibility:
  • Script Integration: Users can access UDS variables directly via the Spyder Variable Explorer to run custom scripts.
  • Custom Algorithms: Users can implement CustomizedAlgorithm functions in specific Python files to integrate new logic directly into the GUI.
  • Module Development: Developers can create new sub-modules by inheriting from the GuiFrame base class.

3. Key Contributions and Algorithms

AngstromPro integrates state-of-the-art algorithms for STM data correction and analysis, moving from traditional real-space methods to optimized Fourier-space techniques.

• Background Subtraction:
  • Implements 2D Plane Subtraction (fitting a 2D polynomial surface) and Line-by-Line Subtraction (fitting 1D polynomials per scan line) using least-squares fitting to remove sample tilt and scanner bow.
• Perfect Lattice Correction:
  • Corrects thermal drift and piezoelectric non-linearity by transforming the real-space image into a perfect lattice.
  • Uses a 2x2 affine transformation matrix derived from reciprocal lattice vectors (Bragg peaks) selected in the FFT.
  • Solves for shear ($b_y$) and scaling ($s_y$) factors based on lattice geometry (e.g., hexagonal vs. square) to enforce specific angles (e.g., 60° or 90°) and equal vector magnitudes.
• 2D Lock-in Technique:
  • Extracts local amplitude and phase maps of specific momentum components ($G_i$) corresponding to Bragg or CDW peaks.
  • Optimized by performing the convolution in Fourier space (multiplying by a Gaussian window in reciprocal space) rather than real space, significantly increasing computational speed.
• Lawler–Fujita (LF) Correction:
  • Corrects picometer-scale distortions caused by piezo creep and temperature drift.
  • Uses the 2D Lock-in technique to extract position-dependent phase shifts ($\theta_i(r)$).
  • Calculates a displacement field $u(r)$ to map the distorted experimental lattice back to an ideal reference lattice, effectively "unwarping" the image.
• Sub-Atomic Precision Registration:
  • Aligns multiple scans taken under different conditions using a 3x3 affine mapping matrix.
  • Requires three pairs of corresponding points to solve for rotation, shear, scaling, and translation parameters, enabling precise image multiplication or division.

4. Results and Validation

The authors validated AngstromPro using a NbSe$_2$ lattice topography and simulated data:

• Processing Efficiency: The software successfully handled massive N-dimensional datasets, demonstrating the ability to perform background subtraction, perfect lattice correction, and LF correction in a streamlined workflow.
• Accuracy:
  • Perfect Lattice: Transformed distorted hexagonal lattices into perfect hexagons, with Bragg peaks forming a perfect hexagon in reciprocal space.
  • LF Correction: Successfully sharpened diffuse Bragg peaks in the FFT, indicating the removal of spatial phase distortions.
  • Registration: Achieved a cross-correlation maximum of 0.966 (vs. 0.855 for unregistered images) after aligning two scans, demonstrating sub-atomic precision.
• Visualization: The dual-panel interface allowed for real-time visualization of FFTs and synchronized manipulation, significantly reducing the time required for feature selection and analysis.

5. Significance

• Unified Platform: AngstromPro consolidates fragmented workflows into a single, cohesive environment, bridging the gap between data acquisition, management, and advanced analysis.
• Reproducibility: By automatically recording the processing history for every UDS variable, the software ensures that complex analysis pipelines are fully traceable and reproducible.
• Community-Driven Development: As an open-source project with a modular architecture, it lowers the barrier for researchers to contribute custom algorithms and modules, fostering a collaborative ecosystem for Scanning Probe Microscopy (SPM).
• Future-Proofing: The N-dimensional data handling and extensible design prepare the field for future advancements in STM, including autonomous experiments and tip manipulation.

In summary, AngstromPro represents a significant leap forward in STM data analysis software, offering a robust, transparent, and highly efficient toolset that addresses the specific needs of modern condensed matter physics research.