Probe Field Manual

How to Photograph a Scene for Analysis

Simple capture habits that make scene analysis clearer, more useful, and easier to defend later.

Start wide before you move in

The first shot should show the whole scene. Wide context tells Probe how objects relate to each other and prevents the analysis from losing orientation. If you begin with a tight detail, the report may understand the object but miss why its placement matters.

After the wide shot, take medium and close images of the most important details. The best documentation set moves from context to detail, not the other way around. Think of the wide image as the map, the medium image as the neighborhood, and the close image as the specific evidence you want reviewed.

Build a simple photo sequence

A useful sequence does not need to be complicated. It needs to answer the questions a reviewer will have later: where is this, what changed, which detail matters, and what else was nearby? Probe can analyze a single upload, but a stronger case file often starts with a deliberate set of source images.

Establishing shotCapture the whole room, vehicle, doorway, package, work area, or outdoor scene before touching anything.
Relationship shotShow the important object with nearby surfaces, walls, labels, fixtures, or reference points.
Detail shotMove close enough for cracks, marks, labels, residue, edges, or small components to be readable.
Alternate anglePhotograph from a second position when depth, obstruction, reflection, or shadow could change the interpretation.

Prioritize light and stability

Blur and darkness remove evidence. If you can, turn on lights, steady the phone, and avoid shooting while walking or twisting around the frame.

Use available lightBright, even light helps edges, text, surfaces, and small objects read correctly.
Hold for one more beatA stable frame usually matters more than taking the shot instantly.
Avoid aggressive zoomMoving closer often beats digital zoom because it preserves detail.

Show scale and position

A clear scene photo should answer where things are, not just what they are. Include nearby surfaces, walls, floor lines, door frames, package edges, lane markings, or landmarks so placement is obvious. The goal is not artistic framing; the goal is enough visual context for a reviewer to understand location and relationship.

If a particular object matters, take one shot that shows it in context and another that isolates the detail. Avoid placing your hand, shoe, or improvised object into the scene for scale unless that is normal for your documentation process, because extra objects can become confusing signals in the later report.

Take extra angles when the scene matters

One image can produce a useful read, but multiple angles reduce ambiguity. A second or third perspective often clarifies whether something is open, damaged, leaning, blocked, reflective, or simply hidden from the first camera position.

This matters most for accidents, property condition, delivery damage, maintenance issues, move-in records, workplace documentation, or any case where you may want a defensible record later. If an angle changes the apparent story, that is useful information. Upload the clearest image for the first report, and keep the supporting images with your case materials.

Before you upload to the web platform

The web platform works best when the source image is easy to inspect on a larger screen. Before uploading, pause for a quick quality check. If the scene is important, this step saves time because it prevents you from building a report around a photo that was doomed by blur, glare, or missing context.

Check sharpnessZoom into the original photo and confirm the detail you care about is readable.
Check edgesMake sure the key object is not clipped by the frame or hidden behind foreground clutter.
Check lightingRetake the photo if bright glare or deep shadow is hiding the evidence area.
Check contextIf the image shows damage but not location, add a wider supporting frame to your record.

Common mistakes to avoid

Cropping too tightlyYou lose the layout and force the analysis to infer context that is not visible.
Shooting through clutterForeground obstructions confuse both human review and AI interpretation.
Relying on one dramatic angleInteresting composition is not the same as useful documentation.
Skipping follow-up detail shotsIf a broken edge, label, or spill matters, capture it clearly on its own.
Editing before analysisHeavy filters, markup, compression, or screenshots can remove details that the original photo still had.

How better capture improves the report

Probe uses the image to generate a narrative, deductions, confidence signals, and annotated overlays. Each of those layers depends on the visibility of the underlying evidence. A wide, sharp, evenly lit image can support stronger object relationships and cleaner overlays. A cropped or dark image may still produce a report, but it will usually leave more uncertainty for you to review.

Good capture does not make the analysis certain. It simply gives the web platform better material to organize. The final review still belongs to the person comparing the output against the source image, the context, and any follow-up evidence.

Keep notes outside the frame

If context matters, write it down in your own notes instead of trying to force every explanation into the photo. A timestamp, location note, order number, maintenance ticket, or short description of what changed can help later review without cluttering the image itself. Probe can organize visible evidence, but your outside context may explain why the scene matters.

Field note: If you are unsure, take one more photo. Extra context is usually more valuable than trying to make one perfect shot do everything.

Related Pages

Open the Case File

Every scene tells a story.

Run evidence analyses on the web platform, then generate visual evidence reports from completed analyses. 3 free monthly analysis credits, with case history, deductions, overlays, and timeline output.