[Erha Image Recognition Artificial Intelligence Vision Sensor] 3. General Settings and Face Recognition

annysky2012

[Erha Image Recognition Artificial Intelligence Vision Sensor] 3. General Settings and Face Recognition [Copy link]

 

After the Erha's firmware upgrade is completed, normal operations can be performed.

Erha supports switching between Chinese and English. For someone like me whose English is not very good, I still feel more comfortable with Chinese.

The default language of Erha's menu interface is English. To set the language of the menu interface to Chinese, please follow the steps below:

(1) Move the Function Button to the right until “General Settings” is displayed at the top of the screen.

(2) Short-press or long-press the "Function Button" to enter the secondary menu parameter setting interface of "General Settings".

(3) Toggle the "Function Button" to the right, select the "Language" parameter (the last one), then short press the "Function Button", then toggle the "Function Button" to the right again to select "Simplified Chinese", then short press the "Function Button", and HuskyLens will automatically switch to Chinese language.

After the settings are completed, select face recognition in the main interface and you can identify the avatar. Face recognition can detect any facial contour and can identify and track learned faces.

After searching for some face portraits on the computer, aim Erha at the area with faces, and the screen will automatically select all detected faces with white frames and display the word "face" respectively. If the "+" in the center of the screen is not aimed at any face frame, the RGB light on the other side will not light up.

Aim the "+" in the center of the Erha screen at the face you want to learn, and short press the "learn button" to complete the learning. If the same face is recognized, a blue box will appear on the screen and display "Face: ID1". This means that face recognition is now available.

Erha has a very powerful function which is continuous learning. If the recognized portrait changes angle, Erha will not be able to recognize it. In order to enable Erha to recognize portraits at any angle, you can use Erha's continuous learning function. The learning method is:

Aim the "+" in the center of the Erha screen at the face you want to learn, and press and hold the "learning button". A yellow frame will be displayed on the screen with the label "Face: ID1", indicating that Erha is learning the face. Then, aim the yellow frame in the center of the Erha screen at different angles of the same person's face, such as the front face, the side face (or multiple photos of the same person), and enter the various angles of this face. During the learning process, the RGB light is yellow. Then, release the "learning button" to end the learning. If Erha recognizes a learned face, the face will be selected by a blue frame on the screen and "Face: ID1" will be displayed.

If Erha detects a learned face, it will select the face with a blue frame and mark it as "Face: ID1". The frame size will change with the size of the face and it will automatically track the face. At this time, the RGB indicator light is green.

If you want to delete the avatar, in the image recognition interface, short press the "learning button" and the screen will prompt "Press again to forget!". Before the countdown ends, short press the "learning button" again to delete the last learning. The "+" will be displayed in the center of the screen, indicating that Erha is ready to learn new things.

深圳小花

Niu Niu Niu

Fred_1977

Very cool, I want to know the principle of face recognition? Including the hardware principle, the principle of software implementation, can you briefly explain it?

annysky2012

Fred_1977 posted on 2021-12-28 19:31 Very impressive. I want to know the principle of face recognition? Including the hardware principle and the principle of software implementation. Can you briefly explain it?

The specific principle has not been introduced in any official materials or documents.

The hardware is just a camera and a CPU, and we have no idea about the principles of the software.