IngenicT23+sc530aiBoard
Project: Board Introduction | page_id: 1340
| Ingenic T23+sc530ai Intelligent Recognition Module | |
|---|---|
| CPU | 64KB + 128KB L1 / L2 Cache |
| SENSOR | SC530ai |
| Memory | Built-in 512Mb/1Gb |
| System | Linux |
| Algorithm | Supports deep learning algorithms with high accuracy and flexibility, human shape detection, face detection and face recognition, and supports cry detection, vehicle detection and pet detection |
| Boot Time | Less than 3 seconds |
| Video Encoding | H.264/H.265/MJPEG, resolution up to 2592*1920@30fps |
| Audio Input | Supports multiple sampling rates: 8K/12K/16K/24K/32K/44.1K/48K/96K |
| Network | RJ45 10/100M |
| Starlight-Level ISP | Special optimization for low-illumination video surveillance scenarios, new generation 2D/3D noise reduction, sharpening enhancement, ROI-AE, advanced WDR, DRC, distortion correction |
| Security | AES/RSA/SHA/TRNG/OTP, supports secure boot |
| Package | 22nm process technology; package: QFN88/BGA192 |
| Operating Temperature | 0 to 70℃ |
| Service Life | Continuous operation life of more than 5 years |
Product Introduction
Board Size: 32*32mm
Chip Specifications
- T31 Description
CPU
• Ultra-high main frequency, up to 1.5GHz.
• 1.5GHz SIMD128 AI acceleration engine
• 64KB + 128KB L1 / L2 Cache
• Risc V independent small core
Video Encoder
• H.264/H.265/MJPEG encoder
• Resolution up to 2592*1920@30fps
• World-class encoding engine
• Supports multiple streams and multi-parameter configuration
Starlight-Level ISP
• Special optimization for low-illumination video surveillance scenarios
• New generation 2D/3D noise reduction
• Sharpening enhancement, ROI-AE
• Advanced WDR, DRC
• Distortion correction
Memory
• Built-in DDR: 512Mb/1Gb
Security
• AES/RSA/SHA/TRNG/OTP, supports secure boot
AI Fast Implementation
• Supports deep learning algorithms with high accuracy and good flexibility
• Supports human detection, face detection and face recognition
• Supports cry detection, vehicle detection and pet detection
Package
• 22nm process technology
• Package: QFN88/BGA192
Fast Boot
• Supports fast boot
• Dual boot
• Fast auto exposure / auto white balance
• Stable image output in about 200ms
Extended Circuits
• Supports 4-channel digital MIC array
• Supports IoT-WIFI / BT / 4
• Supports SLCD display
• Supports UVC/UAC
Audio
• Internal audio codec
• Supports multiple sampling rates: 8K/12K/16K/24K/32K/44.1K/48K/96K
• Supports I2S
• Supports echo cancellation
Peripheral Circuits
• Supports WDT, ADC, UART, I2C, SPI, GPIO, SDIO, PWM, USB, GMAC
Interface Circuits
Power
Debug Serial Port
12V Power / USB Interface
IRCUT Interface
Hard Photosensitive Interface
bootsel0
SMB1 Interface
12V Power / RS232 / 485 Interface
FAQ
| Question | Answer |
|---|---|
| 1. Does this come with a camera? | Yes, it is a pure module and comes with a sensor, but without a lens |
| 2. Is the source code for license plate recognition provided? | The source code is not provided |
| 3. Can it run at low power, i.e. wake up to perform recognition when the radar detects a car? | No, it is not that fast. It needs to stay powered on, with a power consumption of about 1W |
| 4. How long does it take to boot? | About 5s; with fast boot optimization it can be about 1s |
| 7. Are there SDK development materials? | An HTTP interface document is provided |
| 8. Can the algorithm be customized? | Yes, but the algorithm is charged separately |
| 9. Is the hardware source code open? | No, it is not open source |