Free application: Erha image recognition artificial intelligence vision sensor (using K210, 1000 times faster than STM32H743)

EEWORLD社区

Free application: Erha image recognition artificial intelligence vision sensor (using K210, 1000 times faster than STM32H743) [Copy link]

 

Development board model: Gravity HUSKYLENS AI camera (5 sets in total)

From: DFRobot

Market price: 379 yuan

HuskyLens is an easy-to-use AI vision sensor that uses the new generation AI chip Kendryte K210 . When running neural network algorithms, its performance is 1000 times faster than STM32H743 . It has built-in machine learning technology and has functions such as face recognition, object recognition, object tracking, color recognition, line patrol, and QR code label recognition . It comes with a 2.0-inch IPS display, and no computer assistance is required for parameter adjustment . The debugging process and recognition results are directly displayed on the screen, and what you see is what you get. Its onboard UART / I2C port can be seamlessly connected to mainstream controllers such as Arduino, micro:bit, LattePanda, Raspberry Pi, etc., and is widely used in AI education, STEAM education and maker fields.

How to Apply

>>Click here to fill out the application form and submit your evaluation sharing plan (evaluation period: December 12-January 16) . You will have a chance to get this kit for free.

Test data

Video Introduction

Can it be understood as the engineering version of Xiaomi camera? What else can this AI camera do? | HUSKYLENS artificial intelligence vision sensor

Features: Erha Image Recognition is simple but not "simple"

One-click learning: With a very smart design, the onboard AI chip has built-in multiple algorithms. You only need to press the learning button to let HuskyLens learn new things.
Machine learning: The built-in machine learning technology enables it to recognize faces and objects, which is smarter and more powerful than ordinary sensors.
Built-in screen: It comes with a 2.0-inch IPS display, and no computer assistance is required for parameter adjustment. The debugging process and recognition results are directly displayed on the screen. What you see is what you get, which is very convenient.
Extreme performance: It uses the new generation AI chip Kendryte K210, which performs 1000 times faster than STM32H743 when running neural network algorithms .

Technical specifications

· Processor: Kendryte K210
· Image sensor: SEN0305 HuskyLens: OV2640 (200W pixels), SEN0336 HuskyLens PRO: OV5640 (500W pixels)
· Supply voltage: 3.3~5.0V
· Current consumption: 320mA@3.3V, 230mA@5.0V (current values are typical values; face recognition mode; 80% backlight brightness; fill light off)
· Connection interface: UART, I2C
· Display: 2.0-inch IPS, resolution 320*240
· Built-in functions: face recognition, object tracking, object recognition, line tracking, color recognition, label recognition, object classification
· Size: 52mm*44.5mm

Detailed introduction and development tutorials, etc.

>>Click here to enter

For more event information, >>Click here to go to the event page

annysky2012

Any chance to play it?

nmg

annysky2012 posted on 2021-11-25 16:36 Can I have a chance to play it?

If you write a good application plan, you will have a chance. It is clearly reminded in the application: Don't just write "company selection, study a little bit" this kind of application. I just went to the backend to see that there are still netizens who are happy to write like this. I don't know why

For such an application, the judges don’t know what content you can share after the application is accepted, so they will cut it out directly.

annysky2012

nmg posted on 2021-11-26 09:34 annysky2012 posted on 2021-11-25 16:36 Can I have a chance to play? If you write a good application plan, you will have a chance. It is all clear in the application...