Dusun IoT supports custom embedded mainboard solution for AI Camera based on Rockchip RK3588 SoC.
In the AI era, traditional cameras offer only basic surveillance functions and cannot meet the digital needs of commercial and governmental sectors. They fall short in data analysis, real-time monitoring, and intelligent alerting. Equipping these cameras with a powerful “brain” can revolutionize efficiency across various industries.
The Dusun IoT embedded AI camera board solution uses the Rockchip RK3588 chip, which features an octa-core processor (4 Cortex-A76 and 4 Cortex-A55) and demonstrates its powerful computing capabilities, flexible multimedia processing functions, rich interfaces and expandability, efficient AI processing capabilities, robust communication abilities, and low-power design in AI camera applications.
This board is for reference only and can be customized according to needs.
The functionality of AI cameras, such as temperature measurement, license plate recognition, and face recognition, relies on the execution of different algorithm models and data processing, which places varying demands on the camera’s computing power. Lack of computing capability leads to functional failures or reduced accuracy. The Dusun RK3588 embedded board for AI Camera features Rockchip’s self-developed tri-core NPU, capable of collaborative or independent operation, providing up to 6 TOPS computing power. This is sufficient to support algorithms such as facial recognition, object detection, and behavior analysis, enabling swift and accurate intelligent analysis.
Introducing INT4 / INT8 / INT16 / FP16 mixed operations, enabling network model conversion based on frameworks such as TensorFlow / MXNet / PyTorch / Caffe
The RK3588 development board solution supports 8K@60fps H.265/VP9 video decoding and 8K@30fps H.265/H.264 video encoding, with support for simultaneous encoding and decoding, enabling a maximum of 32 channels of 1080P@30fps decoding and 16 channels of 1080P@30fps encoding. The powerful video encoding and decoding capabilities can deliver 8K high-definition images with more refined picture quality.
The embedded mainboard is equipped with a MIPI CAMERA interface, supporting the integration of mainstream industrial / commercial cameras. It easily handles high-pixel requirements and supports high-frame-rate real-time image and video processing, reducing intermediate data steps and improving response speed.
The onboard infrared night vision illumination can provide powerful lighting effects even in the dark or harsh weather conditions, ensuring clear monitoring images and reducing blurring and noise.
In the era of artificial intelligence, more devices need to be stably networked. The Dusun RK3588 AI camera mainboard is equipped with a Gigabit Ethernet interface and supports an LTE CAT4 module, as well as LTE and GPS antennas, ensuring the stability of data transmission and providing stable connectivity for each connected device.
Dusun provides SDK and BSP of embedded mainboard tailored for AI camera applications, so partners only need to focus on application development without the need to port underlying drivers. We has mature experience in the hardware design of embedded AI camera mainboards and can provide professional technical support. We offer a full range of services, from embedded mainboard design, development to manufacturing.
The RK3588 SoM has a wide range of application scenarios, which can be applied to smart cockpits, smart big screens, commercial displays, edge computing, IPCs, NVRs, high-end tablets, ARM PCs, AI cameras, AGV intelligent robots, and other terminals.
