Nuvoton.AI

NuMaker-HMI-MA35D1-S1

NuMaker-HMI-M467

NuMaker-IoT-M467

1. Object Detection

Example: Automatic Light Intensity Adjustment

Use a camera to collect image data of a room.
The Cortex-A35 processes the data from the camera to detect objects (e.g., people, furniture) in the room.
Adjust the brightness and direction of the lighting system automatically based on the number and location of the detected objects to provide the best lighting effect.

2. Biometric Recognition

Example: Facial Recognition Lighting System
Use a camera to capture the faces of people entering a room.
The Cortex-A35 runs a facial recognition algorithm to identify the identities of individuals in the room.
Adjust the lighting color, brightness, or mode automatically based on the preferences of the identified individuals.

3. Object Classification

Example: Smart Lighting Security Surveillance
A camera continuously monitors a lit area.
The Cortex-A35 processes the images to classify objects as "safe" or "potentially dangerous" (e.g., regular household objects vs. unauthorized intruders).
The system can trigger an alarm or notify security personnel if a potential hazard is detected.

4. Real-time Recognition

Use multiple sensors (e.g., light, motion, temperature sensor) to collect data about the lighting environment. Implement an anomaly detection machine learning model. The Cortex-M4 microcontroller continuously monitors the sensor data. Once an anomaly is detected, the system can take appropriate actions, such as adjusting the lighting, sounding an alarm, or notifying maintenance personnel. 

NuMaker-M55M1

1. Anomaly Detection

Use the M55M1's DSP and neural network accelerator to implement environmental anomaly detection, such as smoke, abnormal temperature, or unauthorized entry.
Use the built-in analog comparator and temperature sensor to monitor real-time environmental changes. When an anomaly is detected, send an alert through the connected communication interface.

2. Object Detection

Utilize the M55M1 development board's high-speed processing capabilities and built-in neural processing unit for efficient object detection, such as identifying people, animals, or vehicles.
Combine with an external camera to perform real-time object detection through image analysis technology and adjust the lighting intensity or trigger security alarms based on the detection results.

3. Biometric Recognition

Use the M55M1 development board to process biometric data like facial or fingerprint recognition to provide more personalized lighting settings.
With advanced machine learning algorithms, fast and accurate biometric recognition can be achieved and used for personalized lighting settings or increased security.

4. Object Classification

Use the M55M1's machine learning capabilities to automatically classify objects and activity types in a room, enabling intelligent lighting control.
For example, the system can recognize whether someone is reading, watching TV, or performing other activities and automatically adjust the lighting to suit different scenarios.

5. Real-time Recognition

The M55M1 development board's powerful computing capabilities can be used for real-time recognition of environmental changes, such as the increase or decrease in the number of people in a room, to adjust the lighting to save energy automatically.
In addition, the real-time recognition function can also be used for smart building management, such as space utilization efficiency analysis.