The world's first human-eye camera solves long-distance shooting face puzzles
On October 20th, Green Shen pupil released a human eye camera based on the human eye imaging principle in Beijing, which is used to solve the problem that common surveillance cameras on the market cannot clearly photograph distant human faces.
Face recognition technology has always been widely used in fields such as security monitoring, banking finance, and national defense science and technology, relying on its non-aggressive, human-computer information communicability. Under the support of the depth algorithm, the recognition accuracy of the face recognition technology even exceeds the human eye, but due to the low quality of the face captured by the front-end device, the actual scope of the face recognition is limited to within four or five meters. Geling deep pupil CTO Zhao Yong believes that five meters away from the camera, even in the high-definition picture, the face of the pixel is only 75 * 75pixel below, but in order to achieve good recognition results, the pixel should be 100 * 100pixel or more.
To solve this problem, many manufacturers have introduced 4K or 7K ultra-high-definition surveillance cameras. Although the pixels are high, they are expensive, low in frame rate, and require high bandwidth. The most important point is that they want to capture distant objects. Because of the breadth of sacrifice, it has not been widely promoted. Another idea is that this human eye camera developed by Greene Hitomi can be seen far and wide, based on the working principle of the human eye.
Greenshen pupils use the patented technology of dynamic instantaneous pixel allocation to instantly increase the effective pixels of partial images by more than 100 times. The overall image can reach 200 million equivalent pixels, clearly identifying human faces within 50 meters, and seeing the whole body within 100 meters. Features, vehicle information within 200 meters. In addition, the camera has a field of view of 70Â°, which is more than one hundred times faster than the existing face-taking camera with a viewing angle of only 5Â°-6Â°.
Of course, information capture is an aspect. On the other hand, it requires front-end local real-time algorithms. The camera incorporates the Nvidia TEGRAKI chip, which can output structured video streams or data information in real time after loading optimized image processing algorithms.
In addition, it is worth mentioning that the network bandwidth occupied by mainstream HD cameras is basically the same, and there is no need to upgrade the existing infrastructure. And the integration is simple, and it can be connected to the system platforms of most manufacturers, such as the mainstream video network monitoring and management platform, NVR and so on. In addition to supporting its own face and vehicle identification system and providing an integrated solution, the deep-pitch human eye camera is also compatible with mainstream third-party face recognition and vehicle identification systems.
According to Zhao Yong, the deep-eye human eye camera is mainly used in the field of security monitoring that the company has been cultivating. The target customers are mainly public security systems and traffic police. At present, the product is in the pre-sales testing stage. The specificity of the customer leads to centralized purchasing at the end of the year. Therefore, there are no orders yet, but it has been given to a number of units for testing and the products will be shipped at the end of the year. Although Zhao Yong did not directly give the price, he said that the price was lower than the price of some 4K monitored 20 or so.
Faced with the increasingly fierce competition of face recognition technology companies in the field of security monitoring, like Face, Image, whether in the recognition rate or the application gap is also narrowing, the increasingly converging advantages and disadvantages of the algorithm, how to rely on differences to seize the larger Market share is the question of Green's deep pupil thinking.
The advantage of Green's deep pupil is that it has a hardware development team and hopes to get a single breakthrough from the camera's personal face recognition. Deepin Hitomi also has great expectations for the camera, Green Hitomi said: "This product is our new product, but also belongs to the rank of killer. From next year, it will be the most important source of our sales revenue."
From the current point of view, human eye cameras are also the first in the world, performance indicators leading a number of existing products, but the cost is much more expensive than traditional cameras, there are still some difficulties in the promotion of popularization.
JBL, JBT, JB-TL series specific Parallel Groove Clamp is suitable for unendure force splicing and branch on the aerial wire, being used with insulation cover, It`s action is insulation protection.
â—† Select oxidize-resistance aluminum alloy
â—† All parts connected with each other, it can`t l;lose parts during installation.
â—† An are embrance the most part,conductor can`t change.
Insulation cover performance feature:
â—† Withstand voltage: â‰¥ 18kV keep voltage one minute not breakdown
â—† Insulation resistance: ï¼ž 1.0x1014Î©
â—† Environment temperature: -30â„ƒ~90â„ƒ
â—† Weather proof performance: have good performanced after 1008 hours
artifical climte aging test.
If you have any questions, please contact with us directly.
Our company is solemnly committed to the majority of customers: reasonable price, short production cycle, considerate service. We look forward to cooperating with you.
Parallel Groove Clamp
Parallel Groove Clamp,Jby Parallel Groove Clamp,Copper Parallel Groove Clamp,Aluminium Parallel Groove Clamp
Shandong Vantage International Trade Co., Ltd. , https://www.steelplate.be