With the development of machine vision, artificial intelligence, and human-computer interaction, high-intelligence robots in science fiction movies are no longer out of reach, and consumer-grade robots are gradually appearing in the supermarkets and families, and they have undertaken some physical labor for human beings. However, in recent days, some netizens have exposed the mentally retarded behavior of their own sweeping robots, which has caused a lot of netizens to discuss the current application of robots. It is conceivable that once the own robots are exposed to various strange events due to technical problems, it is unbearable for the enterprise. It is necessary to recall the product maintenance, product line transformation, and the damage of the corporate brand. Trust is greatly reduced. However, from the attention of a large number of netizens and the media, the huge market potential of robots has been gradually realized. It is imperative for robot companies to choose machine vision technology that adapts their products to improve product performance and optimize user experience. In the field of robotics, the current mainstream implementation of intelligent recognition on 2D images through algorithms, but because the 2D image itself contains limited information, even if the algorithm is advanced, the input information will become an intelligent short board, if there can be comprehensive three-dimensional information, each The three-dimensional contours and physical features of the objects will be more fully recognized, and the AI ​​application capabilities such as navigation, trajectory, and recognition will be improved. Selecting 3D depth camera as machine vision technology, performing SLAM construction, implementing obstacle avoidance, following targets, human-computer interaction, automatic grabbing, etc., can meet product performance while taking into account cost, and is recognized by the industry as a cost-effective solution. There are currently three mainstream 3D depth camera technologies : 1) Structured Light: The structured light projects specific optical information onto the surface of the object and is captured by the camera. The information such as the position and depth of the object is calculated according to the change of the optical signal caused by the object, thereby restoring the entire three-dimensional space. 2) TOF (TIme Of Flight): Captures the time of flight of the near-infrared light from transmission to reception through a proprietary sensor to determine the distance of the object. 3) Stereo System: Use two-shot to shoot an object, and then calculate the object distance by the triangle principle. As can be seen from the above figure, the advantage of structured light is that the technology is mature, the depth image resolution can be made relatively high, and the lower hardware cost can be used to achieve or exceed the effects of the other two 3D camera technologies, and in algorithm development. Not too high, and with the upgrade of technology research and development, such as anti-light level is also constantly improving. From the international giants Apple, Microsoft, and Intel have chosen the application of structured light technology to their own products, the industry has more certain advantages of institutional optical technology! In the case that PremeSense is no longer licensed by Apple, and Intel RealSense products are expensive, only a company in the country can provide cost-effective, mass-produced structured light technology solutions. Its 3D depth camera Astra, Astra Pro, Astra mini can realize the SLAM construction of the robot, realize obstacle avoidance, follow the target, human-computer interaction, automatic capture and other functions. Click to view the Obi Vision Robot Solution Video (https://v.qq.com/x/page/y0540sz3r2v.html) Obi Vision Robot Solution Demo The Astra series depth camera is mainly composed of an infrared camera, a laser and an RGB camera. The size is also two. The module size is 165mm*38mm*30mm. The detection range is 0.6-8 meters. The measurement accuracy is within ±1m. 1-3mm; viewing angle 58.4°; resolution can be up to 1280*1024@7FPS. The size of the Astra mini is 80mm*20mm*19.3mm, and the detection range and accuracy are the same, but the relative volume is smaller and the application is more. The core 3D computing chip, depth algorithm, and system support SDK are also completely developed by the company. Since its establishment, the company has applied for nearly 300 patents. The R&D team has the technical strength to support from the underlying chips, deep algorithms, systems, frameworks, and upper-layer applications, so it can achieve lower-cost consumer-grade solutions compared to foreign-based chips. product. Recently, Obi Zhongguang will launch a new generation of structured light 3D sensor camera products, which is more compact than the first generation products, with higher resolution parameters and lower power consumption. It will be suitable for robots, mobile phones and tablets. Manufacturers, such as notebooks and other small-scale products that have 3D visual needs, are bound to have a huge impact on the design of smart terminals at this stage! With the continued speech of Obi LIGHT in the field of structured light 3D sensing technology, 3D sensor technology has been applied to somatosensory TV, robot, security and other industries. Obisoft is a global artificial intelligence visual faucet combining software and hardware technologies. Enterprises will strive to make all the terminals in the world understand the world as their vision. Cable Trunking Fittings,Cable Trunking Accessories,Electric Cable Trunking,Pvc Trunking Accessories FOSHAN SHUNDE LANGLI HARDWARE ELECTRICAL CO.LTD , https://www.langliplastic.com
