Create a task from a photo
Upload a photo taken in the field — Sentinel reads its GPS location and suggests the nearby trails.
Someone sends you a photo of a problem on a trail without telling you exactly where it is. If the photo was taken with location services on, Sentinel can find the spot on its own and suggest the right trail.
It is the same task creation as usual — you simply start from the photo instead of the map.
Where: left navigation bar, under General → Tasks, then the chevron on the Create a task button.
Opening the flow
In the Tasks section, the create button has two actions. Clicking it directly starts the normal flow; the chevron opens a menu.
The create button and its dropdown menu.
- 1Clicking the button directly starts the normal flow, the one that begins with the map.
- 2The chevron opens the menu. Choose "Task from an image".
The creation panel opens on the right, on the Upload image step.
Uploading the photo
The upload step, before a photo is chosen.
- 1Drag and drop the photo onto the area, or click to browse your files.
- 2The back arrow and the cross close the panel or step back, without losing anything.
Common camera and phone formats are accepted, including HEIC from iPhones. One photo at a time: this is the one used to locate the task. You can add more later, in the form.
What Sentinel reads from the photo
As soon as the photo loads, Sentinel analyses it and shows the result.
The analysis result: location found and trails detected.
- 1The uploaded photo. Click it to enlarge; the cross on the right removes it.
- 2The analysis result: the GPS location read from the photo, its coordinates, and how many trails were found around it.
- 3"Continue" moves to the next step.
Sentinel looks for trails within 100 metres of the photo's location, with a 20-metre tolerance either side of the path. Three cases:
One trail found. Sentinel keeps that trail and the position projected onto its path. The map placement step is skipped: you go straight to the form.
Several trails found. The next step shows a map with a photo marker where the picture was taken and one marker per candidate trail. You pick the right one. If none fits, the Place on map button hands control back so you can point yourself.
No trail, or no location in the photo. You place the marker yourself on the map. If the photo did have a location, the map is already centred on it and a photo marker gives you a reference point.
Why a photo sometimes has no location: location services were off on the device, or the photo went through a messaging app that strips metadata while compressing it. In that case, ask for the original file — by email or file sharing rather than by message.
Choose the type, then fill in
The next step is the task type: fallen tree, erosion, broken plank, and any types you have added yourself. "Segment" is not offered here — a photo gives a point, not a length of trail. For a segment task, use the normal flow.
Then comes the usual form: name, description, priority, status, assignee, project, estimated time, materials. The photo is already attached — no need to upload it a second time. It goes to the server once the task is created.
Good to know
- You can step back at any time with the arrow at the top: the photo stays in memory, and a Continue button lets you pick up where you left off.
- Removing the photo with the cross also clears the location and the detected trails. You start again from an empty drop area.
- The coordinates shown are the photo's, not the task's. The task's final position is the one on the trail you kept, or the point you placed.