Skip to main content

GPURaster operations reference

Overview​

@luma.gl/experimental/gpu-raster provides optional, graph-native WebGPU operations for two-dimensional scientific and geospatial rasters. Applications control image decoding, coordinate transforms, source buffers and textures, output allocation, command submission, and optional readback; an explicitly selected residency cache can upload and own source buffers on their behalf.

GPURasterTileReader validates application-owned, asynchronously decoded raster sources without choosing a network transport, image codec, or GPU uploader. GPURasterTileCache optionally adds explicitly budgeted CPU/GPU tile residency, cancellation-safe loading, and reusable compiled graphs around that reader. GPURasterTileHaloAssembler coordinates cumulative receptive-field planning and fence-safe neighboring tile leases; GPURasterTileHaloFill and GPURasterTileCoreExtract explicitly assemble native samples and publish seam-safe owned cores through the caller's command graph. GPURasterOverview and GPURasterCategoricalOverview separately generate nodata-aware analytical overview values and coverage without mistaking source-selected samples or presentation mipmaps for scientific reductions. Explicit global initialization, tiled statistic merges, stable-domain histogram replay, and bounded percentile selection preserve caller-owned dataset-wide results across separately processed tiles. The reader, cache, and assembler never submit commands.

Current contributors cover raster metadata, explicit texture/buffer conversion, calibrated pointwise band math, normalized difference vegetation index (NDVI), validity-aware histograms, scalar summaries, contrast/gamma/equalization transforms, analytical thresholds, mask-aware neighborhood stencils, direct convolution, separable Gaussian/box smoothing, signed Sobel/Scharr/Laplacian derivatives, gradient magnitude, binary/grayscale dilation, erosion, opening, closing, deterministic connected foreground components, bounded dense region labels and counts, grouped intensity and spatial region measurements, GPU-resident marching-squares contours, indirect vector overlays, and adapter-limit planning. They implement GPUCommandGraphContributor structurally: calling addToGraph(graph) only declares work. No contributor submits commands or reads results back.

Choose an operation family​

FamilyUse it for
Storage and residencyValidity, decoded sources, tile caches, cancellation, halos, and owned cores.
Analytical overviewsContinuous and categorical multiresolution summaries.
StatisticsHistograms, scalar summaries, tiled replay, and quantiles.
Pixel operationsBand math, NDVI, contrast, gamma, equalization, and thresholds.
Filters and morphologyNeighborhood filters, convolution, gradients, dilation, erosion, opening, and closing.
Regions and contoursConnected components, region measurements, marching squares, and indirect overlays.

How to use this reference​

Start with GPURaster concepts if nodata, validity masks, tile cores, halos, or analytical overviews are unfamiliar. Each operation-family page states ownership, capacity, failure, and performance behavior alongside its API examples.