MUHAMMAD HUSNAIN / FIELD NOTES
Practical explanations of hyperspectral imaging, fine-grained learning and AI systems. Diagrams, tested examples and links to the original research.
Foundations
Learn how hyperspectral imaging measures spectra, builds data cubes and balances wavelength range, resolution and calibration, with a Python example.
Read article · 6 min
Foundations
Compare NIR spectroscopy with hyperspectral imaging, learn reliable chemometric validation and calculate prediction errors with a tested Python example.
Read article · 6 min
Foundations
Compare hyperspectral capture methods, match sensors and illumination, plan calibration and calculate line spacing and motion blur with Python.
Read article · 6 min
Higher-order
Understand graphs, hypergraphs and topology for hyperspectral images, with a tested example and practical guidance on relationships and evaluation.
Read article · 6 min
Weak supervision
Compare image tags, points, scribbles and noisy labels for hyperspectral learning, with a tested spatial-overlap check and clear evaluation guidance.
Read article · 6 min
Foundation models
Explore what foundation models mean for HSI, how SpectralGPT and HyperSIGMA differ, and how to check spectral inputs and evaluate transfer.
Read article · 6 min
Agentic AI
Models, agents, tools and evaluation: a practical guide to bounded, reproducible AI assistants for hyperspectral research.
Read article · 6 min
Tooling
Calculate HSI cube memory, match CPU and GPU work, and use public Python, NumPy, Rasterio, GDAL, PyTorch and Spectral Python tools.
Read article · 6 min
Conference Notes
A verified CVPR 2025 proceedings digest covering HyperFree, HSI calibration, selective fusion and large-image remote-sensing evaluation.
Read article · 6 min
Fine-grained analysis
See how smoothing erases tiny HSI objects with interactive maps, a tested Python example and metrics that reveal what overall accuracy misses.
Read article · 5 min
Fine-grained analysis
Explore open HSI problems in mixed pixels, rare classes, boundaries, sensor shift and evaluation, with a practical checklist and public references.
Read article · 5 minCONFERENCE NOTES
Scoped proceedings summaries that connect published contributions with their evidence, limitations and practical value.
THE READING PLAN
Published guides are linked. Planned topics are a roadmap, not unfinished articles.
| Article | Status |
|---|---|
| Hyperspectral imaging fundamentals: spectra, pixels and data cubes | Published |
| NIR spectroscopy and hyperspectral imaging: what changes? | Published |
| Capturing hyperspectral data: sensors, scanning and calibration | Published |
| From radiance to reflectance: a practical calibration guide | Planned |
| HSI preprocessing in Python: bad bands, noise and scaling | Planned |
| Spectral libraries and material identification: promises and limits | Planned |
| Article | Status |
|---|---|
| Higher-order representations for hyperspectral learning | Published |
| Weak supervision in hyperspectral learning | Published |
| Foundation models for hyperspectral imaging | Published |
| Self-supervised HSI learning: objectives and evaluation | Planned |
| CNNs, Transformers and Mamba for HSI: a fair comparison | Planned |
| Adapting spectral foundation models to a new sensor | Planned |
| Article | Status |
|---|---|
| Why smoothing can erase tiny objects in hyperspectral maps | Published |
| Open problems in fine-grained hyperspectral analysis | Published |
| Mixed pixels: when classification needs unmixing | Planned |
| Spatial splits and leakage in hyperspectral evaluation | Planned |
| Beyond overall accuracy: measuring fine-grained HSI quality | Planned |
| Sensor shift, uncertainty and rare classes in HSI | Planned |
| Article | Status |
|---|---|
| The era of agentic AI: tools, control and reliable workflows | Published |
| Hardware and software tooling for HSI analysis | Published |
| Build a reproducible HSI analysis pipeline in Python | Planned |
| Processing large HSI cubes without exhausting memory | Planned |
| AI agents for literature review: sources, checks and limits | Planned |
| HSI experiment tracking: configurations, artefacts and reproducibility | Planned |
| Article | Status |
|---|---|
| CVPR proceedings: remote sensing papers worth reading | Published |
| IGARSS proceedings: spectral sensing and learning research notes | Planned |
| NeurIPS proceedings: geospatial learning and evaluation notes | Planned |
| ICCV proceedings: Earth observation and representation learning | Planned |
| Remote sensing workshop notes: datasets, methods and open questions | Planned |
| From a conference paper to a reproducible HSI experiment | Planned |