Advantage
* Nova MSD 300 sending card and Nova mrv328 receiving card
* Cabinet size:640x1920mm
* Kinglight/Nationstar LED Lamp, Refresh rate:1920-3840hz
* High brightness up to 2000cd/sq.m, even in the sunlight conditions can see clear, but low power consumption to save the electric power cost.
* High debugging brightness and no damage to gray scale, achieving the debugging technology for nice image.
* Passed the TÃœV,FCC,ROHS,CE cetification.
Poster LED Display,Poster Led Display Screen,Mirrior Led Display,Led Poster Display Guangzhou Chengwen Photoelectric Technology co.,ltd , https://www.cwleddisplay.com
1. Utra Thin and Light weight, Cabinet weight only 19KG and 86mm thickness.
2. In direct sunlight, the outdoor LED Display can be high-definition and high-brightness, high refresh rate, high contrast
3. Can withstand high and low temperature, can be used in extremely harsh environment
4. Modular design, Front and Rear maintenance system available, Separate and exchangeable power and data unit can be easily removed for easy maintenance
5. Outdoor IP65 is waterproof, dustproof and corrosion proof, ensures smooth and reliable operation under variable and extreme weather conditions.
Jun Zheng: Open the Blue Ocean Strategy, T01 or Achievement 2018 Face Recognition Explosives
With the rapid rise of artificial intelligence, the market has begun to embrace both AI and intelligent voice technologies. Developing an AI chip that offers optimal performance, low power consumption, and cost efficiency has become a crucial factor in enabling AI applications. This chip is not only the core of AI "energy" but also the driving force behind its explosive growth. Today, beyond smart speakers, machine vision is making waves in industries such as security. The commercialization of machine vision is now taking shape, with numerous face recognition products based on this technology being launched.
At the 16th China International Social Public Safety Expo (referred to as "AMB"), machine vision products for facial recognition became the main attraction. Technologies like face capturing and recognition were highly popular, with companies like Hikvision and Fluorite showcasing their latest innovations, including the Fluorite C5S wall-mounted smart camera and Ai Mi's Yunzhi AI camera. Products from Qinglian Internet, Wanjiaan, Wangwang, and Weiwei also made appearances, all featuring deep learning capabilities, multi-target detection, 1080P resolution, and real-time video at 30 frames per second. These features are powered by Junzheng’s first-generation machine vision chip, T01.
T01 marks Junzheng’s beginning in the field of machine vision — a true "from 0 to 1" breakthrough. As camera resolutions increase and the need for faster linkage grows, front-end chips have struggled to meet these demands. Despite years of development, low-cost, real-time face recognition cameras remained out of reach. However, recent releases from Love Eyes and Fluorite at the AMB have changed this, breaking through the limitations of front-end computing power and speed, and gaining significant industry attention.
The image above shows the T01 chip, which is set to revolutionize AI applications by addressing key pain points. It was showcased in face recognition cameras at the 16th AMB, highlighting its role in driving technological innovation. Junzheng's vision of "from 0 to 1" innovation is central to its mission, and T01 represents the starting point of this journey.
Let’s take a closer look at the basic specifications of T01:
- ★ XBurst CPU
- ★ CV Engine V1.0
- ★ Multi-input/output interface support
- ★ Host interface
- ★ Typical power consumption of 700mW
In terms of core performance, T01 offers:
- ★ Simultaneous detection of up to 16 objects
- ★ Support for detecting more than 30 targets at once
- ★ Multi-resolution configuration, supporting real-time monitoring up to 1080p
- ★ Common data input/output interface support
- ★ 30 frames per second real-time detection rate
Interestingly, T01 does not include an ISP or H.264/H.265, and it has minimal external interfaces. That’s intentional — T01’s sole purpose is high-quality target detection. As Junzheng’s first machine vision chip, it combines the XBurst CPU, inherited from Jun’s traditional CPU technology, and the CV Engine, the company’s first dedicated machine vision ASIC engine.
Unlike common AP SOC or MCU chips, T01 is a significant step forward in the machine vision domain. It marks the start of Junzheng’s journey in this area and sets the stage for future developments.
T01’s superior performance and clear positioning make it ideal for large-scale deployment. While it can be applied across various industries, its primary focus should be on human-related attributes — such as face, half-body, and full-body detection. On T01, the single-target detection accuracy exceeds 97%, with a speed of 30 people per second and 30 frames simultaneously. Detection distances can reach up to 30 meters under normal conditions, and over 100 meters for full-body humanoid detection.
One of T01’s key advantages is its ability to perform multi-object hybrid cross-checking, significantly reducing the chance of missed detections. Additionally, its real-time detection and precise angle analysis capabilities enhance face recognition quality. T01 can capture high-quality frontal faces from video streams, improving cloud-based recognition results.
Thanks to T01’s speed and angle tracking features, application scenarios can be greatly expanded, including crowd counting and behavior analysis. These unique features enable partners like Aurora and Fluorite to innovate more effectively. By integrating T01 as a coprocessor, developers can reduce engineering complexity and streamline product development, making it easier to implement functional improvements.