◆Sample program available & technical information released◆ Recognizing target objects from the entire camera image (AI/Image Recognition)

We will introduce a method for object recognition that does not cut out the object from the camera image and uses a single image as training data!
For more details, please see the URL below or related links.
https://qiita.com/shimamotosan/items/1a0c6c73b1e60ae5db43
■ What is DOBOT Magician?
Despite being priced around 150,000 yen, it is a small robotic arm equipped with a 4-axis arm and comes with standard tool heads such as a gripper, suction cup kit, lighting & drawing, and 3D printer.
The dedicated software allows programming by arranging blocks, enabling beginners to smoothly learn and operate the robotic arm. The main unit does not require prior safety training, making it easy to install and use in various locations and scenarios. Afrel provides original guides (control learning texts) using dedicated software (DobotStudio) and programming languages to support the utilization of the robotic arm.
【DOBOT Magician explanatory article】
https://learninglab.afrel.co.jp/1074/

Inquiry about this news
Contact Us OnlineMore Details & Registration
Details & Registration
Related Links
This is the introduction page for DOBOT Magician.
We will introduce use cases for the robotic arm.
When distinguishing objects, we use OpenCV to learn from the cropped target objects in the camera image or input them into a trained model for recognition. Therefore, fine-tuning is necessary to successfully recognize and crop the target objects from the camera image.
This time, I would like to modify the program to use a single image as training data without cropping the target object from the camera image, enabling classification of the target object.
Related product
Related catalog(9)
![[Presentation of Materials] Understanding the 9 Features of the Small Robot Arm DOBOT Magician! Live Broadcast of the Product Introduction Seminar](https://image.mono.ipros.com/public/catalog/image/01/2ef/581655/IPROS27524302385260153207.jpeg?w=120&h=170)