Simon entrance audit
Last updated
Was this helpful?
Last updated
Was this helpful?
Step 1: Access the Job Ticket on Asana
Locate the Job Ticket:
Log into the Asana platform to access the assigned job ticket for the configuration audit.
Review the ticket details, which include:
Specific instructions for the audit.
Location and cameras to be reviewed.
The date and time range for the footage to be analyzed.
Take notes of all essential details, such as the date, time, and specific cameras, to ensure accuracy during execution.
Step 2: Access Dragonfruit Platform
Login and Organization Selection:
Log into the Dragonfruit AI platform.
Select the appropriate organization from the list (e.g., Simon for the Simon Entrance Audit).
Navigate to the Location Tab:
Go to the Location tab within the platform to locate the cameras specified in the job ticket.
Select the required cameras.
Choose Video Footage:
Locate and access footage from the specific date and time range mentioned in the Asana ticket (e.g., October 16th).
Step 3: Manual Review of Footage
Observe and Count:
For each minute of the footage, count the number of people crossing the region (indicated by a line filter on the camera).
This region represents a boundary where every crossing event should be noted.
Capture Screenshots (If Required):
If the instructions specify, take screenshots of people crossing the region as evidence.
Ensure the screenshots are clear and correspond to the timestamps.
Note down the manual count of individuals crossing the region for the entire time range specified
Step 4: Generate AI Insights
Create an Insight Report:
Navigate to the Insights section on the Dragonfruit platform.
Create a new insight report by selecting:
The same camera and location previously analyzed.
The date and time range matching the manual review.
Review the AI Report:
The platform generates a bar graph and data details showing the AI-detected number of people crossing the line.
Note these values carefully for comparison.
Step 5: Record Results
Internal Config Audit Sheet:
Log the manually counted numbers and AI-detected numbers into the Internal Config Audit Sheet.
Attach screenshots if required to support the audit findings.
Accuracy Analysis:
Compare the manual counts against the AI results to evaluate the AI model's accuracy in detecting people crossing the line.