SPECTRAshows camera detection, confidence, model review and a verifiable proof result.

SPECTRAEvidence vision

Multispectral security / evidence AI

SPECTRA

A working browser proof that translates camera image into demonstrable detection, measurable quality limits and human verification.

Working research proof

The case in 30 seconds

SPECTRA

Problem: A working browser proof that translates camera image into demonstrable detection, measurable quality limits and human verification.
Solution: From your own camera image to local detection, measurable image quality, level of evidence and exportable audit record.
Role and focus: Night vision, evidence AI
Status: Working evidence-vision proof
Result: Case, product site, live lab

Live lab / try it yourself

Test what camera image really proves

Use your own or sample image, compare detection with image quality and see when the system shows evidence or deliberately does not draw a conclusion.

Open the vision lab View the story

01 / InputCamera image and quality limitYou can change or choose this yourself

02 / SystemLocal vision AI and verificationComputer vision · ONNX inference · Evidence scoring

03 / EvidenceA measurable level of evidenceThis makes the operation visible

InputOwn camera image

AnalysisLocal object detection

EvidenceQuality + source framework

BorderDon't make up an identity

The design question

When does image become evidence?

A sharp live feed is not automatically useful for a safety decision. The task determines what counts: detecting someone requires a different image than recognizing clothing or confirming an identity.

Detect

Is there demonstrably a person or vehicle present?

lowest image demand

Observe

Are direction, action and context visible?

more pixels + time

Recognize

Are distinguishing features reliably observable?

detail + good lighting

Identify

Can an identity be confirmed with an appropriate method and basis?

separate validation

Reproducible baseline measurement

One modality is missing

Ten pre-selected night moments, one fixed detector and one pre-determined boundary show why a second sensor is relevant.

Fixed fusion test10/10

presence reports at the ten monitored times

View method and limits →

Visible9/10

A visible model misses one dark moment.

Infrared with RGB model5/10

The same AI is not automatically suitable for a different sensor.

Score fusion10/10

The highest sensor score captures the complementary miss.

This is a fixed regression test from one ongoing public survey series, not a claim of 100% market accuracy. It is precisely that boundary that remains visible.

High-end system direction

Sensors with a task

Not one magical color, but a coordinated chain in which each sensor does what it is demonstrably good at.

LWIR

Thermal finds

Detection in darkness and difficult background, without identity claim.

850 / 940 NM

NIR follows

Monochrome detail and texture with a custom trained model.

VISIBLE

White preserves color

Brief, controlled evidence recording for clothing, vehicle and context.

EVIDENCE GATE

AI limits

Measure sharpness, target size, exposure and uncertainty before making a conclusion.

Nighttime sample image for theSPECTRAbrowser analysis

person66%

Live in the browser

Test your own frame

Choose a JPG, PNG or WebP. The model runs locally in the browser, drawing real detections and simultaneously assessing light, contrast, sharpness and target size.

No upload to a server
Original image remains visible
Detection is not identity
Insufficient evidence remains a valid outcome

Open the live lab →

Design principles

Evidence before effect

SPECTRAdoes not make uncertainty more beautiful, but useful. Every conclusion remains linked to source pixels, measurement values and a human follow-up step.

Bronvast