With the rapid development of the Internet of Things industry, the number of connected IoT devices globally increased from 5.2 billion in 2015 to 12.6 billion in 2020, and is expected to reach 24.6 billion in 2025. In this case, the demand for personalized and lightweight segmentation scenarios will become more and more vigorous, and the intelligent application of AIoT will be gradually enriched. The more lightweight the application scenario, the more need to improve the efficiency of the platform, chip and hardware devices, and reduce the deployment difficulty and cost. Intelligent operation can be achieved without relying on the interaction with the cloud.

In addition, the massive demand for segmentation also requires the product to have a higher cost performance, and can fully benefit from old, reduce unnecessary waste of resources.

Lightweight edge computing applications, reducing deployment costs by 30%

TC-RV1126 development board series products are based on the RV1126 platform of Rockchip. It has the characteristics of high performance, low power consumption, lightweight deployment and more cost saving. It is very suitable for edge computing applications in small and medium-sized scenarios. Compared with the previous generation TC-RK3399 development board, the deployment cost of TC-RV1126 development board series products is reduced by 30%, which is a rare cost-effective edge computing hardware selection.

Equipped with vision chip, high performance and low power consumption

TC-RV1126 development board series products is equipped with Rockchip RV1126, which uses quad-core Arm Cortex A7 32-bit kernel architecture and integrates NEON and FPU. Support TensorFlow/MXNet/PyTorch/Caffe and a series of deep learning framework, rich resources and easy development, meet the demand of siege lion different development. The platform has core technologies in neural network, image processing, artificial intelligence, deep learning, SOC system power optimization, face detection, image acquisition, adaptive adjustment of image technology and other fields.

Built-in independent NPU, AI efficiency is higher

It is generally believed that the higher the NPU calculation power, the better. Is this really the case? In fact, high computing power will also bring higher power consumption and construction costs. In light application scenarios, high utilization of computing power but low utilization of computing power is a waste of resources. TC-RV1126 development board series products has an independent built-in NPU, which provides 2TOPS computing power and supports INT8/INT16 mixed operations. It fully meets the application in lightweight scenarios. Besides, it performs AI computing independently, which is more efficient, accurate and less false positives.

Supports a variety of video codec, suitable for different display devices

TC-RV1126 DEVELOPMENT BOARD SERIES PRODUCTS support H.265/H.264/MJPEG video codec, supports multi-level video quality configuration and coding complexity Settings, supports 4K@30fps+1080p@30fps video coding, maximum decoding 4K@30fps, and supports the same encoding and solution; It can adapt the original display devices with different configurations, and combine the excellent codec ability with edge computing. While improving the image quality, the video stream delay is lower, the real-time performance is stronger, and the bandwidth occupation and storage space are reduced, further reducing the application cost.

Rich extension interface, reduce the difficulty of development

TC-RV1126 development board series products are equipped with Ethernet, HDMI out, RS232, RS485, Line in/out, Wi-Fi, TF and other interfaces to meet the requirements of different scenarios. With low power consumption and high integrated circuit design, it can operate smoothly and stably in the outdoor working environment of -40~70℃, and has strong adaptability to harsh environment.

It is applied to the intelligent construction site to achieve efficient and intelligent supervision

TC-RV1126 development board series products have been maturely applied in the field of smart security, face recognition, door lock. TC-RV1126 development board series products built in a variety of site scene algorithms, real-time monitoring and identification of a series of site violations and alarm;

The edge computing products of Thinkcore technology help the intelligent production and management of the construction industry with rich data types, vertical segmentation intelligent analysis and standard open API interface. In the future, Thinkcore Technology will continue to make efforts in the field of edge computing, and introduce more software and hardware integrated edge computing solutions that are widely used in all walks of life.