Overview
With the promotion of remote work due to the COVID-19 pandemic, a new way of working has begun to take root. However, this has raised concerns about information leakage risks and the need to establish new business processes. There are no clear measures in place to prevent acts such as employees taking photos of their work screens with smartphones or third parties viewing sensitive information, leading to information outflows. As a result, some departments and job types are unable to work from home. Deep Percept for remote work creates a safe remote work environment using unique AI-based motion analysis technology to detect fraud while protecting employee privacy.
Features
Detection of Unauthorized Activities
An AI with motion analysis capabilities constantly monitors the user’s movements in front of the webcam, ensuring that “secret photography” of the PC screen with smartphones, digital cameras, or feature phones, as well as “peeping” or “impersonation” by third parties, is not overlooked.
Immediate Blocking of Unauthorized Activities
- If an action is detected as an attempt to photograph with smartphones, digital cameras, or feature phones, screen filters are activated to hide the content and guard against photography.
- If it is determined that there is ‘use by a third party’ or ‘use near the person,’ screen filters are activated to make the content invisible.
- When an activity is detected, the surrounding situation is photographed, the image is encrypted, and the management system is notified.
- The content displayed on the screen at the time of detection is screen-captured, the image is encrypted, and the management system is notified.
- Detailed settings for alerts and stops at the time of detection are possible.
Recording of Unauthorized Activities
Unauthorized activities captured by the webcam and the information displayed on the PC screen at that time are encrypted and stored on the management server, and reported to the administrator.
Consideration for Privacy
- Analysis of images captured by the webcam is performed only by AI, and images other than those detected as unauthorized cannot be viewed by third parties, including administrators.
- Settings are available to blur parts of the image other than the face and camera, such as the interior of the room.
Functions
Peeping Detection
Detects peeping from reflections of unauthorized individuals.
Peeping Photography Detection
Detects photography activities by nearby smartphones, digital cameras, or feature phones.
Personal Verification
Verifies that the PC user is the person who was registered beforehand.
Third-Party Detection
Detects if the PC user is not the person who was registered beforehand.
Absence Detection
Blocks the screen if the PC user moves away from the PC.
Attendance Management
Prevents overwork by allowing administrators to grasp the attendance status of employees.
Position Monitor
Provides a visualizer function to ensure the user’s face fits within the PC camera’s angle of view, reducing unnecessary detections.
Edge Computing
AI operates on the PC, which (1) reduces communication between the cloud and the PC, (2) does not unnecessarily upload information to the cloud, and (3) allows monitoring to continue offline even during wide-area failures in the cloud.
Product Image
For those considering remote work for employees, click here:
For those considering a home call center, click here:
Service and Support
Based on the requests and feedback from corporate users, this service undergoes several functional additions and improvements annually, updating all AI models, agents, and web management apps.
A trial version of the agent is also available for customers considering implementation.