1. Product Introduction The vertical face recognition management module with card swiping adopts the Rockchip RK3288 processor, equipped with an industrial-grade dual-camera and live face recognition technology, making it a high-performance and highly reliable product. It supports 1:1 and 1:N face comparison and retrieval, as well as identification of ID cards, IC cards, etc. It … Read more
RKNN is the model type used by the Rockchip npu platform, with model files ending in .rknn. Rockchip provides a complete model conversion Python tool, making it convenient for users to convert their self-developed algorithm models into RKNN models. Additionally, Rockchip offers C/C++ and Python API interfaces. Use RKNN-Toolkit2 on a computer to complete model … Read more
Lattice recently cited data from IHS at its new product launch event, indicating that the market growth rate for FPGA chips will outpace the overall semiconductor market over the next 4-6 years. While different research institutions provide varying estimates for the FPGA market size, this is indeed a consensus. Typically, these reports mention potential drivers … Read more
1. Product Overview 1.1 Applicable Scope KJD.RK3588.D belongs to a smart live broadcast motherboard, widely applicable to: live broadcast platforms, smart self-service terminals, smart retail terminals, industrial control hosts, and robotic devices. 1.2 Product Overview The KJD.RK3588.D series adopts Rockchip’s new generation flagship high-end processor RK3588. The RK3588 uses an 8nm process technology, featuring an … Read more
1. Chip Parameters 1. CPU 1.1 XBurst®-1 Core – XBurst® FPU instruction set supports IEEE754 compatible single and double precision formats – XBurst® 9-stage pipeline microarchitecture, operating frequency of 1.5GHz · MMU – 32-entry unified TLB – 8-entry instruction TLB – 8-entry data TLB · L1 Cache – 32kB instruction cache – 32kB data cache … Read more
Introduction to T40 CPU 22nm CMOS low power BGA356, 14mm x 14mm x 1.22mm Dual Core 1.2GHz AI Engine 8TOPS Real-time H.256/H.264 encoding capabilities: 3840×2160@30fps Support up to 3 sensors Support MIPI and DVP interface sensor Support maximum resolution 3840×2160 • XBurst2 up to 1.2GHz, Dual Core, Dual-issue, high performance and low power implementation of … Read more
1. Core Functions Supports 4-channel vehicle AHD camera input encoding and synchronous recording. Supports driver face recognition and abnormal alarm upload. Equipped with 8T computing power, supports various AI detection algorithms, extensible to the ADAS field. Supports multiple video streaming protocols including hls/RTSP/RTMP/flv. Supports national standard protocols such as GB28181/JT808, extensible to departmental standard protocols. … Read more
Function Parameters SigmaStar Low Power Series Chip AI Performance Comparison SigmaStar Low Power Series Chip AI Performance Comparison (Part II) Sigmastar & Junzheng Junzheng T40 NOR Fast AI Camera Junzheng A1 Full HD Network Video Decoder Junzheng A1 33-channel (Dual Ethernet) Streaming Device Junzheng T31ZX + GC4653 4MP PTZ Fixed Focus Camera Junzheng T40 + … Read more
1. Product Overview 1.1 Application Scope of Junzheng A1 Function Board The Junzheng A1 is an embedded NVR function board that uses standard DC12V power supply, supports dual SATA hardware interfaces, and is widely applicable to smart NVRs, video decoders, soft routers, and related products. 1.2 Product Overview The Junzheng A1 features a 1.4GHz dual-core … Read more
AI Revelation Written by | Exploration Studio (Please indicate the source when reprinting) 🛠️👩💻 AI Data Centers May Face Power Crisis by 2027 There arebenefitsat the end With the rapid development of Artificial Intelligence (AI) technology, its electricity demand is also continuously increasing. Recent research reports indicate that AI data centers may face serious power … Read more