![[Technical Information] Recognizing the target object from the entire camera image (AI)](https://image.mono.ipros.com/public/product/image/a61/2000729799/IPROS31540842792200048537.png?w=280&h=280)
[Technical Information] Recognizing the target object from the entire camera image (AI)
Sample program available ■ Introducing a method for object recognition that uses "a single image as training data without cropping the object from the camera image"!
Here is a case article introducing the tabletop-sized small robot arm, DOBOT Magician. ■ Article link: https://qiita.com/shimamotosan/items/d3a06e834d332b914973 ■ Content When distinguishing objects, we use OpenCV to take the extracted objects from camera images as training data or input them into a trained model for classification. Therefore, fine-tuning is necessary to effectively recognize and extract objects from camera images. This time, we will modify the program to classify objects using a single image as training data without extracting objects from the camera image. Keywords: Python, OpenCV, AI, image recognition, dobot ■ About DOBOT Magician DOBOT Magician is a robot arm equipped with a 4-axis arm, priced around 150,000 yen. The device does not require prior safety training, making it suitable for various locations and applicable in a wide range of scenarios. * DOBOT Magician product introduction page https://afrel.co.jp/product/dobot
- Company：アフレル 東京ロボティクス本部
- Price：100,000 yen-500,000 yen
![[Video Introduction] Features of DOBOT: Direct Teaching Edition](https://image.mono.ipros.com/public/product/image/ed1/2000639273/IPROS30183062897805175699.png?w=280&h=280)
