Child face distance alert system

A non-transitory machine-readable medium having instructions stored therein for monitoring distances between a human face and a display screen of a smart device, the instructions, when executed by a processor, causing the processor to perform operations, the operations comprising:

• capturing, using a camera of the smart device, an image frame of a human face of a user operating the smart device;
• classifying, using a trained neural network model running on the smart device, the human face on the image frame to determine whether the human face is a child face, the trained neural network model determining a first rectangle having a fixed size;
• in response to determining that the human face is a child face, creating a second rectangle by extending a bounding box around the human face, the second rectangle proportional to the first rectangle;
• determining, using the trained neural network model, a distance of the human face to the display screen based on a ratio between a diagonal of the first rectangle and a diagonal of the second rectangle;
• comparing the distance of the determined child face to the display screen with a predetermined distance to determine whether the determined child face is within the predetermined distance to the display screen of the smart device;
• and
• in response to determining that the determined child face is within the predetermined distance, generating an alert message on the display screen of the smart device.

A data processing system, comprising:

• a processor;
• and
• a memory coupled to the processor to store instructions for monitoring distances between a human face and a display screen of a smart device, the instructions,
• when executed by a processor, causing the processor to perform operations, the operations comprising:
• capturing, using a camera of the smart device, an image frame of a human face of a user operating the smart device;
• classifying, using a trained neural network model running on the smart device, the human face on the image frame to determine whether the human face is a child face, the trained neural network model determining a first rectangle having a fixed size;
• in response to determining that the human face is a child face, creating a second rectangle by extending a bounding box around the human face, the second rectangle proportional to the first rectangle;
• determining, using the trained neural network model, a distance of the human face to the display screen based on a ratio between a diagonal of the first rectangle and a diagonal of the second rectangle;
• comparing the distance of the determined child face to the display screen with a predetermined distance to determine whether the determined child face is within the predetermined distance to the display screen of the smart device;
• and
• in response to determining that the determined child face is within the predetermined distance, generating an alert message on the display screen of the smart device.