Image Discovery

Description

Turn scientific images and 3D data into organized, measurable, interpretable knowledge.

Image Discovery is the scientific data organization and analysis environment being developed by Macroscopic Solutions. It is designed to receive structured outputs from Image Remote—or independently added scientific imagery and 3D datasets—and keep the visual evidence, specimen record, calibration, annotations, measurements, interpretations, and reporting context together.

Features · Pricing and license options

View the Rainbow Scarab as a model and a Gaussian splat

The screenshots below show real outputs in the current Image Discovery development build. The same Rainbow Scarab captured through Image Remote can be explored as an Agisoft Metashape mesh or as an appearance-rich Gaussian splat. Both viewing paths are working in Discovery’s 3D workspace.

From automated capture to connected inspection

1. Capture and process in Image Remote. Collect focus stacks across rotary views, stack each view, and retain the source images, processing settings, calibration evidence, and Methods records.
2. Create complementary 3D outputs. Use the Metashape workflow for a textured mesh and the Gaussian-splat workflow for photographic appearance. Each output remains a distinct derivative with its own processing history.
3. Inspect in Image Discovery. Choose a saved output from the model catalog and open its linked Methods record. Researchers can compare the representations without searching through separate output folders.

Scale and measurement with a visible evidence trail

Inspection rulers and scale bars are available for both viewers. Where applicable capture calibration and reconstruction geometry are available, Discovery can display an estimated physical scale. These scarab examples use that estimated path; an image scale bar alone is not an independent verification of 3D dimensions. Ruler exports retain the measurement method and scale basis. Mesh rulers connect picked geometry points; splat rulers connect Gaussian centers and do not establish a solid surface or volume.

Reviewed landmark, length, angle, and selected-area analysis requires an eligible mesh and the appropriate scale evidence. A mesh suitable for viewing is not automatically suitable for every analysis. Metashape remains the photogrammetry processing environment; Discovery provides the connected inspection workspace. Explore Image Remote and its capture demonstration.

Planned annotation, measurement, and comparison tools

• Spatial annotations: place points, lines, regions, boundaries, landmarks, and notes directly on the relevant image or model.
• Calibrated measurements: record lengths, areas, intervals, orientations, counts, and other discipline-appropriate measurements in physical units when calibration is available.
• Structured observations: use field-specific terms and repeatable forms rather than burying interpretations in filenames or unstructured notes.
• Side-by-side comparison: compare specimens, views, treatments, time points, or analytical interpretations while preserving the identity and scale of each source.
• Linked evidence: connect an observation or measurement to the exact location, source image, specimen record, and analyst context that support it.
• Reviewable interpretation: keep boundaries, assignments, classifications, measurements, and notes visible so the researcher can revise or validate the analysis.

Built-in guidance without forcing every discipline into one interface

Image Discovery is being structured around discipline-specific workspaces. Each workspace can present the terminology, measurements, annotation types, metadata fields, and review steps appropriate to that field. Built-in instructions remain beside the data, helping laboratories standardize routine work and making advanced image analysis easier to teach, repeat, and audit.

Planned geoscience analysis workspaces

The geoscience images below are illustrative design previews, not screenshots of the current development build. The modules show the depth of field-specific analysis being developed. They begin with calibrated imagery and traceable specimen information, then present tools suited to the evidence and questions of the discipline.

Mineralogy & petrography

• Trace and edit grain, mineral, pore, vein, and alteration boundaries
• Associate point counts and optical observations with locations in the image
• Record candidate mineral assignments without losing the evidence used to support them
• Compare regions or related sections while retaining scale and sample identity
• Export mapped observations and a reviewable analytical record

Sedimentology & grain analysis

• Review and correct individual grain boundaries
• Measure calibrated grain size and summarize the observed distribution
• Record roundness, sphericity, sorting, and related descriptive properties
• Link grain-level observations to the parent sample and collection context
• Compare samples or intervals using consistent measurement and review criteria

Paleontology & morphometrics

• Maintain catalog, locality, formation, horizon, and stratigraphic context
• Place repeatable anatomical landmarks and calibrated linear or angular measurements
• Compare morphology across specimens, views, ontogenetic stages, or groups
• Keep taxonomic observations and reviewer notes attached to the visual evidence
• Prepare annotated figures and measurement records for collaboration or publication

Drill core & structural geology

• Navigate long-format imagery by calibrated depth or interval
• Mark lithologic contacts, fractures, veins, structures, and sampling locations
• Record orientations, confidence, descriptive attributes, and reviewer comments
• Connect observations across adjacent trays, core runs, or imaging segments
• Keep analytical features linked to borehole, depth, specimen, and acquisition provenance

A platform for additional scientific disciplines

The same connected-record approach is intended to support additional discipline-specific workspaces as Image Discovery develops. Rather than applying generic tools to every subject, each module can organize the measurements and observations that matter to that field.

• Biology and natural history: specimen cataloging, morphology landmarks, trait observations, counts, structures, and comparative collection review.
• Materials and engineering: regions of interest, surface features, defects, fracture or wear observations, calibrated inspection, and sample-to-sample comparison.
• Archaeology and cultural collections: object records, condition features, manufacturing traces, material observations, provenance, and comparative documentation.
• Plant, insect, and zoological research: specimen-level annotations, developmental or morphological features, measurements, and linked views across complex image sets.
• Teaching and shared facilities: guided workspaces, consistent terminology, reviewable student or user observations, and complete handoff records.

Natural-history collections and entomology

For collection teams, Image Discovery is being scoped around the relationship between a specimen, its images, its labels, and its physical location. The proposed workflow can connect whole-drawer views with individual specimen records so a curator can compare imaging sessions and investigate what changed over time.

• Compare drawer or tray images across dates while retaining each capture’s scale, source, and review status.
• Associate catalog numbers and label transcriptions with annotated specimen positions in a drawer image.
• Record cabinet, drawer, tray, and specimen locations alongside moves, loans, and return events.
• Keep collecting locality, origin, taxonomic identification, and determination history connected to the visual record.
• Link measurements and morphological observations to the exact specimen image or 3D view that supports them.

These are development directions, not a released feature list. Collection teams can help prioritize which relationships, permissions, and review steps matter most.

Help define the field modules

What would make a scientific image database, viewer, and measurement workspace useful in your lab? We welcome requests for additional field distinctions beyond mineralogy, petrography, sedimentology, paleontology, drill core, entomology, and broader natural-history collections. Tell us which records must stay linked, what you need to compare, and where your current workflow loses context. Share your requirements with Macroscopic Solutions.

Initial availability will depend on release readiness and the validation of each discipline-specific module.

Reproducibility, review, and reporting

• Capture-to-analysis continuity: preserve Image Remote camera settings, calibration, scale information, processing history, and draft Methods documentation.
• Transparent analytical records: retain the analyst’s annotations, edits, measurements, classifications, and interpretation notes instead of flattening them into a final image.
• Consistent review: organize work for internal checking, collaborator review, teaching, or publication handoff without rebuilding context manually.
• Export with context: prepare annotated figures, measurements, structured observations, and supporting metadata while preserving the original scientific assets.
• Extensible field modules: add discipline-specific tools without changing the underlying project, specimen, provenance, and calibration framework.

Scientific accuracy and a reviewable Methods record

Image Discovery keeps each measurement beside the source image or model, scale basis, analyst, selected landmarks or boundary, and processing history. A calibrated length, grain size, core interval, area, angle, or 3D ruler is useful only within the limits of its source and calibration. The unusual strength of this connected workflow is that a reviewer can move from an exported measurement back to the specimen and the choices used to produce it, instead of receiving an isolated figure.

When an independently measured reference of the same feature is available, percentage error can be reported as 100 × |image-derived measurement − independent reference| / |independent reference|. Record the reference method and its uncertainty, sample size, scale calibration, and where on the image or model each measurement was made. A nonzero reference is required. Registration residuals and a visible scale bar do not by themselves establish dimensional accuracy. Estimated scale in the current scarab viewers should remain labeled estimated until independently validated.

For a manuscript, retain the camera and optics, illumination, tile or rotary capture plan, focus-stacking and reconstruction settings, source and derivative IDs, calibration target, any geometric correction, analyst edits, and known limits. Discovery is designed to preserve these relationships and export a Methods-ready audit trail; the accuracy of a particular result still must be measured for that specimen and workflow.

How the four programs connect

Image Remote controls and documents capture, then hands focus stacks to Zerene Stacker, ordered overlapping tiles to Image Pan, multi-view photographs to Agisoft Metashape for a textured mesh, or supported multi-view imagery to Image Splat for a Gaussian representation. Discovery brings the reviewed images and 3D derivatives into a searchable, measurement-aware specimen record. Each processing path has its own settings and validation evidence; third-party software is licensed separately.