User Manual, Quickstart Guide, and Video Demos
The following are the User Manuals of the Sentinare sensor (formerly known as the Cypress visual sensor) in different languages. Please go over the user manual to understand the details of the many features and settings of the Sentinare sensor and the App.
The following is a quickstart guide of the Sentinare sensor and the App, including some video demos. All videos are available at our YouTube channel.
1: Download the Sentinare App
To use the Sentinare sensor, you need to install the free Sentinare Activity Sensor (formely known as the Cypress visual sensor) App on your mobile device. The sensor also needs to connect to the Internet via a WiFi network with WPA2-PSK (aka WPS2-Personal) security protocol.
Choose the following download method depending on your mobile device’s platform:
2: Install the Sensor
- The sensor comes with a USB power adapter with 1.5-meter 20 AWG micro USB cable (not shown in figures below). Please do NOT use other adapter or USB cable, as they may not be able to power the sensor properly. Regular AC power extension cord is recommended to connect the USB power adapter to AC power outlet.
- The sensor can be installed with the USB port in the bottom or on the top. In the latter case, you need to turn on the Upside-Down Installation flag in the App. More details can be found in the user manual.
- Find a few potential installation positions that are about 1.5-2.5 meters from the floor, so that when the sensor is tilted down slightly, it can monitor most of the floor for fall detection and the room for action/face recognition. Try not to screw the sensor on the wall before you find the final spot via the Calibration step below, to avoid unnecessary damage to the wall. We will soon release an App that allows updating the background image without doing the time-consuming floor calibration to speed up the process.
- If you want to use face recognition, including the Region of Interest feature, the ideal installation height is around 2m, so that it can recognize people’s front faces.
- The stand/bracket can be mounted on a horizontal surface, on the wall, or on the ceiling. The rotating arm of the bracket can be installed in one of two directions.
- Power up the sensor, and wait for about a minute until the LED starts to flash white, indicating that the sensor is ready to be added to the system. Please refer to the user manual for detailed installation guides.
3: Create a Sentinare Account
- Open the Sentinare App on your mobile device, tap the “Create an Account” on the front page.
- Note: The Sentinare account is different from your shopping account at the AltumView website.
- In the “Sign Up” page, fill in your name, email, and password. The email and password are used for login. The password should have at least 8 characters and include upper case letter, lower case letter, and digit.
- Choose your Server: Tap the pencil icon to choose a server location.
- Currently, there are three servers in Canada, United States, and China.
- When a sensor is sold, its serial number is added to the white list of one server, based on the country of the customer. If you are not in these three countries, please choose the US server.
- Read and accept the Terms of Services, then select “I agree to Terms of Service”.
- Tap the “Sign Up” button. After that, you will receive a verification code in your email account. If you could not get it after a few minutes, please make sure your email is spelled correctly, and check your junk mail folder.
- Enter the verification code to the next “Confirm Email” page, then tap “Continue” to enter your account.
The following is a video demo of creating a Sentinare account.
4: Add the Sensor to Your Account
- Now you can add your sensor to your account by tapping the Rooms icon at the lower left corner of the screen.
- Note: A sensor can only be added to one account at any time. If a sensor has been added into another account, its status will be displayed as “Added”, and it cannot be added to a new account unless it is removed from the original account. See the User Manual for details.
- Note: In the App, each sensor must be added to a Room, and a Room can have up to 10 sensors. When a new account is created, a default room “1st Floor- Room 1” is created.
- To add your new sensor, click on “1st Floor- Room 1” in the Rooms page.
- In the next room page, click the “+Add Device” command. The App will search and list all nearby sensors.
- Select the sensor to be added according to its serial number. The LED of the selected sensor will flash green for 20 seconds. In the next page, create a name for the sensor.
- Setting up WiFi: Next the App will set up the WiFi for the sensor. The App will first scan available WIFI networks with WPA2-PSK (aka WPS2-Personal) security protocol. Selecting your WiFi network and entering its password, the LED on the sensor will first turn to red, then yellow and blue, and eventually become steady green. After that, this sensor will connect to WiFi and the server, and appear in the list of sensors in the room.
- Firmware Upgrade: After adding a sensor to the account, if its firmware is not the latest, please upgrade the sensor’s firmware. Please always use the latest firmware as we are constantly improving the firmware. The upgrade might take a few minutes, and the sensor might reboot multiple times. After the upgrade is finished, the sensor will show a steady green light. Please refer to the Upgrade Firmware section in the User Manual for details.
The following is a video demo of adding a sensor to an account, setting up WiFi, and performing floor calibration.
5: Calibrate the Sensor
- The sensor must be calibrated to detect the floor before it can detect falls correctly. You need to redo the calibration if the angle or position of the sensor is changed. The sensor only sends alert when a person falls to the calibrated floor area. Falling on sofa, couch or bed will not trigger alert.
- To protect people’s privacy, make sure that nobody is in the view of the sensor during the calibration.
- Users must be within Bluetooth range of the sensor (~5m) in order to do the auto-calibration. Manual calibration can be done remotely.
- To do the calibration, tap the “Calibrate” command. If the sensor has not been calibrated before, the App will first show some instructions about the installation height and angle, and will ask users to select if the sensor is installed upside-down or not, and if the background image should be natural mode or privacy mode. After making these choices, press the “Begin Calibration” command will start the calibration.
- For the Cypress sensor, the calibration takes about 15 seconds. For the Sentinare 2 sensor, the calibration takes about 30 seconds.
- To ensure good fall detection performance, please install the sensor’s height and angle can capture a large area of floor. If the floor area is too small, a warning message will be displayed to remind you to adjust the installation height and angle of the sensor.
- To update the calibration, tap the “Recalibrate” command, the sensor will take a new background image and detect the floor in it. Tap the “Save” button to save the result and exit the page.
- You may need to adjust the position of the sensor and redo the calibration several times before finding the optimal spot with good coverage of the room and the floor.
- Manual Calibration: Sometimes the auto-calibration algorithm may mis-calibrate some non-floor areas as floor, or miss some floor areas, which could cause wrong fall detection results. In this case, users can manually modify the calibration result using the four editing tools in the Calibration page: Draw, Rectangle, Erase, and Clear.
The following is a video demo of calibration using the private mode.
6: Add People (Optional)
- Some features of the Sentinare sensor rely on face recognition, such as statistics of each person’s daily activities and face-recognition-based Region of Interest (unless the Detect Anyone option is selected). To enable these features, users need to add the head photos of people that need to be monitored by the sensor.
- Note: Fall detection does NOT need face recognition.
- To add a person, tap on “+ Add Person” at the top of the “People” page. The App first needs to get a head image of the person from the mobile device’s camera or from the image library, as shown below.
- After getting a head image, the “Add Person” page will appear, where the user can input the name of the person and assign this person to a Person Group, such as Seniors or Staff.
- In the next page, users can add or delete photos for the person. In this page, users can change Person Group, and input the person’s detailed information.
For best performance, take the following photos of the head:
- Front view
- 45o left and right views (with both eyes visible)
- 15o top view
The entire head should be in the circle with both eyes visible. The photos should be clear, well-lit, without shadows or other people.
The following is a video demo of adding person to an account.
7. Linking with Alexa Together (Optional)
Sentinare has been proudly selected by Amazon’s Alexa Together (currently in US only) as one of only three fall detection devices, although Sentinare can work independently, and also offers many other features.
Alexa Together is a new senior care service that provides 24/7 hands-free access to an Urgent Response professional emergency helpline, which is especially helpful for seniors living alone.
With the Sentinare sensor and the optional Alexa Together subscription, the Echo devices in the loved one’s Amazon account (not the caregiver’s Amazon account) will ask if urgent response is needed when a fall is detected by the Sentinare sensor. The following is a video demo.
The sensor is now ready to look after your loved ones! It will monitor their activities, collect statistics, and send alerts when emergencies such as falls are detected. You can check their real-time activities, or review their activity history.
The following are some demos of other powerful features of the Sentinare sensor. Please refer to the User Manual for details.
Video Demos of Other Powerful Features
When the Sentinare sensor detects a fall, it will send an alert to the Sentinare app. The alert will include a short clip of stick figure animation to show how the person falls, and can help to identify if it is a real fall or a false alarm. If the name of the person is recognized (via face recognition), the name is also included in the alert and the stick figure clip.
Waving Hand Detection
If a loved one needs help from the caregiver, he/she can wave to the Sentinare sensor to trigger the waving hand detection alerts. The sensor’s LED will flash red for a few seconds, and then turn to purple and send the alert to the App. The alert sensitivity can be adjusted from the App (See the User Manul for details). The video below shows the results with low, medium, and high sensitivities respectively.
Users can define a restricted region, so that the Sentinare sensor can send an alert when persons in the specified groups enter the region, for example, when persons with dementia enter the exit area of the long-term care facility. Face recognition is required, unless “Detect Anyone” is selected.
Stick Figure Recording
The Sentinare sensor can record and play back all human activities in terms of stick figure animations, which provide valuable visual information while protecting the privacy of the person.
Personal Activity Statistics
When the face photos of a person are added to the Sentinare account, the Sentinare sensor can recognize different persons, and collect their individual activity statistics for playback and analysis.
Please note that similar to any other Artificial Intelligence products, the algorithms of the Sentinare sensor are not perfect, although we are working hard to improve them constantly.
By purchasing our products and using our services, you agree to our terms and conditions as stated in our website and in the App, including the disclaimers, limitations of our products and services, and limitation of liability.
In particular, you agree that the device is not guaranteed to detect all possible abnormal events that it is designed for, and is not guaranteed to collect all information without any error. We cannot guarantee in any way that the device will prevent accidents, death or personal injury to you or others, or prevent property damage, illegal entry, or undue delay in any emergency service response, and we are not liable in any manner for any injury, loss or damage resulting therefrom. If you do not agree with these clauses, you should not use our products.