Uncompromising Precision in
Fabric Uniformity
Inspect textile surfaces instantly. Map irregularities, evaluate structural texture via GLCM, calculate uniformity risk thresholds, and generate traceable quality records in real time.
Interactive Analysis Preview
Hover over the grid cells to simulate real-time defect localization. Toggle filters to inspect different vision algorithms.
Scan Analysis
00.00.00Hover cursor over grid cells to inspect sub-grid computer vision metrics.
Core algorithm selection
Calculations merge Standard Deviation and GLCM Contrast/Energy metrics for high-resolution defect localized mapping.
Empower Quality Assurance with Advanced Vision
Engineered for textile manufacturers and industrial materials laboratories to replace subjective manual checks with deterministic data.
Sub-Grid Localization
Divide fabrics into high-resolution sub-grids (up to 20x20) to pinpoint localized structural anomalies with millimeter accuracy.
GLCM Texture Analytics
Analyze Grey Level Co-occurrence Matrices to quantify fabric contrast, energy, homogeneity, and correlation metrics directly.
Deviation Heatmap Overlays
Superimpose real-time heatmaps onto high-resolution fabric scans. Easily toggle value counts, grid frames, and color-coded risk maps.
Traceable Inspection Logs
Store every scan in a persistent SQLite database. Generate shareable, audit-ready reports complete with histograms, parameter values, and raw images.
Success Stories
Hear from materials laboratory leads and textile quality assurance managers who trust our vision-based inspection engine.
"Integrating UniformityAI transformed our lab trials. The GLCM texture analysis identifies micro-density shifts that were previously invisible to the human eye, providing quantitative verification."
"We scan high-tech fabrics for aerospace composites. The sub-grid deviation heatmaps allow us to spot localized grid anomalies instantly and archive ISO 9001 quality certificates automatically."
"The web UI combined with the database storage makes compliance tracking a breeze. We run uvicorn locally for zero-latency offline inspections on our manufacturing floor. Incredibly reliable."
Inspection System Q&A
Everything you need to know about the grid-based inspection calculations and local deployment environment.
The inspection engine divides the fabric image into a user-defined grid. It computes local statistics for each cell—such as standard deviation of pixel values and GLCM (Grey Level Co-occurrence Matrix) features (Energy, Contrast, Homogeneity). The cells are then graded against your chosen threshold, and a combined uniformity score is derived. Deficient cells are highlighted in red or yellow on the heatmap.
The platform supports standard web image formats: JPEG, PNG, and WebP. Thanks to our integrated OpenSeadragon viewer, you can upload high-resolution multi-megapixel industrial fabric scans, and view and zoom in on individual fabric threads without browser performance lag.
While the Next.js app serves the web interface and handles the database, the computational heavy-lifting is performed by a dedicated Python FastAPI service (under services/analysis-service/). It uses OpenCV, NumPy, and scikit-image to execute high-speed matrix convolutions and texture classification.
When scanning fabric samples, the edges of the image often contain background backdrops, mounting frames, or frayed threads. The 'Ignore Edge' slider allows you to specify a boundary thickness (number of cells) to omit from the overall statistics, ensuring edge artifacts do not skew your core material results.
Yes, every uploaded batch and analysis run is stored in a local SQLite database via Prisma. You can review historic runs, compare parameters, reload cached heatmap cell details, and trace quality trends over time directly from the history sidebar.
Ready to Eliminate Surface Quality Variance?
Run automated fabric inspections on-premise or connect to external scanners. Get immediate, deterministic uniformity analytics and traceable compliance archives.