YOODS’ Vision Technology

Home
Robot Vision

Since the advent of digital cameras, image processing technology has become increasingly important in factory automation (FA) processes. Since its founding in 2004, YOODS has focused on this importance and has been developing related technologies. More recently, we have focused on vision technology as a visual function for robots and are concentrating on developing robot vision solutions.

Technology provided by YOODS

type	item
Integration technology	Arm mounting	By attaching a camera to the robot arm, the equipment can be simplified, allowing for workpiece placement over a wide area and compatibility with mobile robots. In addition to the typical fixed position on the robot flange, we also propose attaching the camera to the previous axis (the fifth axis). Attaching it to the fifth axis prevents cable entanglement due to flange rotation.
	Fixed vision	It also supports the relatively common fixed camera system. It is suitable for bag depalletizing where the relationship between the work pallet and the camera remains constant, as well as for correcting the posture of workpieces held by robots and for high-cycle operations.
	off-line	We support technologies that translate logic verification through desktop simulations and offline teaching into real-world applications.
	Automatic Calibration	To use a camera as the “eyes” of a robot, it’s essential to accurately determine the relationship between the robot arm and the camera. YOODS’ calibration system provides a technology that allows for accurate and easy determination of the camera mounting position on the arm.
	Visual Teach	This technology corrects robot movements based on the amount of workpiece movement, and it forms the basis of YOODS’ robot vision platform. We offer a variety of solutions based on this software.
	High precision posture recognition	We provide a technology that uses point cloud data obtained from 3D vision to quickly and accurately determine the position and orientation of a workpiece. Optimizing this function determines the performance of the 3D vision.
Elemental technologies	3D point cloud generation	We provide a comprehensive solution, from hardware to software, for generating 3D point clouds from 2D images. Since the yield and accuracy of the 3D point cloud are affected by the 2D image acquisition conditions, our expertise in 2D image processing plays a crucial role.
	Deep Learning	We offer workpiece recognition technology utilizing deep learning, which has become widely used in the image processing field in recent years. It is particularly effective in segmentation, where specific regions are extracted from the image space, and is used for recognizing workpieces with irregular shapes or irregularly shaped workpieces such as bags.
	Image processing technology	We provide image processing and photography technologies that we have cultivated over the years, which can be used robustly in a variety of environments.
	Point cloud processing technology	We provide optimal point cloud processing technology, including processing time, for 3D point clouds, which have a significantly larger data volume compared to 2D point clouds.
Reliable support	FA specifications	Our cameras are designed for use in factory automation (FA). From the mechanical design stage, our cameras are built with the assumption that they will be used in FA environments, providing hardware with excellent environmental resistance.
	Implementation support	We don’t just sell products; we provide thorough support for both hardware and software until they are fully functional.
	SIer collaboration	We will also actively collaborate with equipment manufacturers that provide solutions to end users.
	After-sales service	We provide post-delivery maintenance support and long-term parts and product supply, which are crucial in factory automation.