

Fasteyes - The Body Temperature & Face Reognition Attendance System Improvement

Background
Fasteyes is now a standard product in the company I worked for. When I first saw the product, it was just a body temperature detection prototype, and my job was to improve and extend its functionality.

Previous

Now

Team
- Project Lead
- UI / UX Designer
My Role
- Peggy Cao (UI / UX designer)
- Chang Ting Wei (Front-end engineer)
- Sam Chen (Back-end engineer)
- Auggie (Image processing engineer)
Collaborate With

Goals
As we can see in the picture above, the previous one was just an early prototype, so there was a lot of room for improvement on its UI & UX.
It was not enough to stand out in the market if the product could only detect human temperature and upload the data online. Therefore, we developed a facial recognition attendance system based on the original functions.
Other optimizations such as improving the accuracy of temperature detection, face detection and face recognition, etc. We also added many new functions such as mask detection, ERP integration, and a device binding mechanism.
Redesign
There are 2 kinds of interface in this product. In this article, I would put emphasis on the management website that I designed.

Fasteyes software for facial detection and recognition.

Management website for both desktop and mobile users. (Designed by me)

Design System


Dashboard Page

Original
In the original version, there were some impractical or non-functional features such as "Connected device list" and "Normal temperature amount list for each device", so they were removed in the revision.


Revised
In the revised version, I also added a date and time picker, and a device multi-select dropdown, so that users can easily filter out the data range they need. Also, the original "Normal temperature" card was replaced by "Abnormal cases today".

History Page

Original


Revised
On the history page, I used the same control bar for users to filter the time and devices. Those settings will remain on every page until the user changes them. In addition, a search bar, a download button, and a picture preview field were added.
New Feature - Attendance System

Fasteyes Software
Together with Peggy, the other UX designer, we made a lot of changes to the Fasteyes software for the attendance system.

Original

Revised
- The grayscale background was replaced by the primary color defined by the color system.
- On the large screen, we found that people would tend to move around instead of standing still, and that made recognition difficult. So we narrowed the camera display size and added some text hints.
- A lot more options were added on the settings page.

Report System
In the early stage of development, the recognition accuracy was not quite good, but the MVP had already been released. Therefore, I designed a reporting system to let users manually correct the misjudged cases, and the back-end system would automatically collect these data for developers to further improve the accuracy.


Other Features
- Attendance Record

- The attendance record page only presents the staff checking history and the earliest and latest time based on the selected time range. This function is especially useful for companies that have 3 switch systems.
- Another hidden customized feature is that we can sync the data to the HR system or ERP that users are using.
- Upload Staff List

On the staff management page, users can upload a staff data list at one time, or just add staff one by one.
The Outcome & Next Steps

The Outcome
- A 70% increase in orders and interested buyers.
- An 80% decrease in complaints about bugs and bad experiences received from customers.
- Attract more IoT projects and custom software development.

Next Steps
We found that customers’ demand for IoT customization is greater than that of simply a body temperature & face recognition system. This doesn't mean that they don't need Fasteyes, but that they need more extensive functions
“An IoT Platform”
The below article is about the IoT platform we built for a tech company, and that platform will also become an extended service based on Fasteyes.
Next Project >>

All rights reserved © 2021 Astrid Lin