Minimal Edge AI Board for Smart City Applications !
Connected to CCTV cameras, it uses AI algorithms to recognize objects in real time and outputs control signals based on the recognition results.
🔹Key Features
AI Object Recognition : Users can set multiple areas on the camera screen and set conditions based on the number of objects within the area. When the conditions are met, an alarm signal is immediately sent.
Alarm Signal Generation : The alarm signal is output as a real-time electrical signal through the GPIO port. This signal can be used to turn on a warning light, output an alarm sound, or be used by a nearby service board. The user interface can be configured to turn the signal on or off for a specified period when an object appears.
Other Output Interfaces : Recognized event data can be transmitted to a remote server via the internet network and stored in the board's internal database, where it can be analyzed as a chart through the user interface.
Object Tracking : The direction of object movement can be tracked and alarms can be controlled based on the direction of movement.
User-Specific Object Recognition : Common recognition objects, such as people and vehicles, can be selected from the library. Users can also train the system to recognize specific recognition objects.
Heat Dissipation System : This system supports a heat dissipation solution that essentially eliminates the need for a cooling fan, ensuring stable and noiseless operation even in harsh weather conditions.
Built-in Router : The board features a multi-port wireless router, eliminating the need for separate network equipment and enabling a minimalist configuration. Wi-Fi cameras can be installed wirelessly, and when necessary, a management laptop can be used for remote Wi-Fi access and management. This allows for management even in locations where maintenance is difficult, such as tall utility poles or manholes.
Board and Case Form Factors : The board is by default available in printed circuit board form, but case-based devices are also available.
🔹Diverse Applications
Experience business innovation with customized solutions tailored to your industry.
Smart City Traffic Control System : Installed in minimal spaces, such as within a utility pole enclosure, within a traffic control system, it can visually perform various control functions based on surrounding conditions.
Danger Zone Alarm System : Sends an alarm signal when an object appears within a polygonal area defined in the camera screen.
Automatic ventilation/lighting control system : Alarm signals can be used to control lighting or ventilation.
🔹How to Use
Connection
Connect the camera to an Ethernet port (or connect wirelessly to Wi-Fi)
Connect the signal output via the GPIO port
Connect the board's power supply
Settings
Connect to ai_settop via Wi-Fi with your laptop and access the AI settop control panel via a browser.
Click the camera icon in the upper right corner of the control panel, then press the search button to search for and select a camera.
Click the area designation icon and click counterclockwise on the screen to designate an area.
Check the "Enable GPIO Output" checkbox on the settings screen. When an object (person) appears in the designated area, an electrical signal will be output through the GPIO connector on the front panel. Specifying a time in milliseconds in the Linger Time field will cause the device to turn on or off for the specified amount of time when an object appears. Checking "Invert Output" will cause the device to turn off when an object appears. When the pin output is turned on, the LED icon on the screen will also light up.
🔹Product Specifications
Input
Up to 9 network cameras
Output
Local database storage, GPIO pinout, server transmission, etc.
Key Features
Zone-based object recognition, electrical signal output
