The Problem

No Existing Application

Only researchers who could work with code could process the data. Our challenge was to design an interface that less technical users could understand, and that technical users could use to more efficiently refine their data processing algorithm.

Wide range of uses

Researchers needed to select a period of time to evaluate and a baseline period for comparison. They could evaluate any timeframe from 1 second up to multiple days and the baseline and evaluation periods could also vary dramatically.

Findings

Further Problem Definition Needed

I met with stakeholders with a list of questions to address the key problems the application needed to solve.

How long are tamper events typically?

Minutes, hours, days, longer?

What happens if there is a tamper event?

What actions need to take place?

Inspiration

Video Editor Timelines

Users needed the ability to view data over a large time period but also zoom in on small increments of time for evaluation. I was inspired by video editing software timeline controls which have the power to adjust the zoom level of the timeline to stretch or condense the content as needed.

Final Designs

Dashboard

Users can view past evaluations, separated by camera or facility, or run a new evaluation.

Dashboard interactions

Hovering over an evaluation will highlight the time period that the evaluation took place and any overlap with detected tamper events.

Previous evaluation info

Clicking the information icon on an evaluation lets the user view a history log and related comments.

Camera selection

Researchers will import files from their computer into the application. If a previously created camera is detected, the user will be prompted to use this camera.

If the incorrect camera is detected, users can select a camera manually or create a new camera.

Set baseline period

For a successful evaluation, a clean baseline period must be used for comparison. Most users will accept the default baseline period set by a more technical user.

Custom baseline period

When selecting a custom baseline period, users can utilize zoom controls to comb through previously processed data to find a clean baseline period. They can also set this baseline as the default for future evaluations.

Set evaluation period

All imported data will be selected by default, but users will have the ability to pick a specific evaluation period if they want to look at a smaller data set.

Viewing results

Potential tamper events will be highlighted on the evaluation results screen. The user will be able to zoom in on the timeline to further inspect potential tamper events and then confirm or dismiss them.

When zooming in further on a timeline, spectrogram data will appear that the user can use to compare against the baseline. Selections from the timelines are also overlayed in the graph.