为了实现多种智能设备之间低带宽高画质的屏幕内容分享，设计了一种基于嵌入式平台大屏幕内容分享系统。该系统由跨平台客户端及基于嵌入式平台的服务器端两部分组成，针对传统屏幕内容分享系统占用带宽大、不适合分享高分辨率视频信息的缺点，提出了一种基于场景识别的码率自适应调整算法，该算法采用H264编码方式，能够根据网络拥塞状况及当前屏幕的应用场景对H264码率进行自适应调整，以达到节约带宽、提高画质的目的[另一方面，服务器端基于嵌入式平台，采用多线程并发设计，将视频的接收、解码、显示分线程独立处理，在提高程序的健壮性的同时降低了程序的耦合性。实验结果表明，该系统支持720 p高分辨率视频信息共享，具有成本低、占用带宽小、实时性高的优点。

In order to share screen among intelligent devices with high quality and low band width, a screen sharing system based on embedded system has been designed and implemented, which consists of a crossplatform client and server. For the purpose of reducing bandwidth and improving screen quality, a selfadaptive method based on scene recognition is proposed, which could recognize the application scene intelligently and adjust H264 code rate according to network congestion state and application context. On the other hand, the server has been deployed on embedded board. For the server side, a solution based on multiple threads has been proposed, which consists of main thread, message receiving thread, decode thread and display thread. Every thread focuses on their own task, which could improve the robustness and coupling dramatically. Experiment results show that the proposed screen sharing system could support screen sharing between PC and embedded display system with 720 p high resolution. And this solution has the advantages including real time, lowbandwidth and low cost.