Smart City
基于FPGA的3D人脸特征传输系统
PR034 »
随着社会的发展，人工智能在人们身边出现的越发频繁，智慧景区以及智慧城市也是当前的人们发展趋势，本设计顺应需求，针对部分场景，如景区中的游客流量、游客、性别等基础信息的统计，或无人超市中人物信息的识别以及保存，并将这些数据备份、发送到云端，并在app中对部分数据进行实时更新，在做数据统计与分析时，利用云端中存储的数据，可以很全面且客观的分析出现状，这些工作靠人力实践是非常耗时耗力的。在必要时，还可调用出对象的信息和图像，对图像进行伪3D显示，大大提高了人们对图像的感知力。基于此类功能以及特性，本设计可广泛推广于安保、有人员统计需求类的场景。
本设计以 DE10-Nano 开发板为平台，结合当前最新推出的 DE10 云套件 RFS 子卡搭建了一款物联网系统，在此基础之上加入摄像头、液晶显示屏以及四片半透明介质，组合构成对象人物信息提取系统和佩伯尔幻像伪全息人物显示系统，并通过在arm核中移植linux系统以及在此之上配置的opencv和openvino环境对获取的图像信息进行处理分析，所得到的数据传回fpga，通过云系统发送到app，实现实时对部分数据的监控，每隔一段时间会发送数据到数据库储存供以随时调用。并可通过调用信息，读取对象人物的信息，选择对象的照片进行全息显示。
Industrial
基于FPGA的多路JPEG-LS无损解压缩
PR035 »
随着相机技术的发展，像素分辨率与帧率都得到了大幅度提高，由此需要处理的图像数据量也随之越来越大。在一些如地理信息、医学影像、卫星侦察领域，因数据传输通道变化不大，所以需要对传输的图像数据进行压缩后传输，并在终端解压缩进行图像还原。而在图像压缩中，JPEG-LS具有显著的优势，它比JPEG2000的算法复杂度更低，比JPEG的压缩性能更强。所以JPEG-LS算法具有硬件易实现，复杂度低，同时具有较好的压缩性能。
基于JPEG-LS的压缩目前多采用硬件实现方案，但其解压缩却多采用软件解码方式，而软解存在CPU资源占用过多、功耗较大以及处理实时性等问题，尤其在多路软解下问题更突出。据此提出本项目“基于FPGA的多路JPEG-LS无损解压缩”以满足多路视频解码的实时性。
项目拟采用多路并行处理结构，对接读取的云端多路图像数据。系统由多路相机、服务器、FPGA DE10-Nano套件以及其他配套电路组成，拟实现对4路1080p@50Hz的视频压缩码流进行实时解码。其中压缩码流与解压码流均通过网络与服务器互联。
该方案在JPEG-LS解压缩的速度上有较大提高，尤其在多路图像数据（高帧率、高像素）的处理时，效果更为显著。通过该项目的实施可以较好体现FPGA的重构性、实时性、低能耗等性能。
Smart City
基于FPGA的老年人防跌倒系统的实现
PR037 »
项目主要以FPGA为核心控制系统，通过三轴加速度传感器实时监测老人的运动状态，当老人发生跌倒后，在一定的时间内老人没有通过按键重置系统后，系统将通过GPS系统获取老人当前位置信息，通过GPRS系统呼叫紧急联系人，并且将位置信息通知短信形式发送给其紧急联系人。
Other: Robot
Example: AlphaBot
PR001 »
Robotic era is approaching.
In order to rapidly control the robot, we hope to use the FPGA fast and easily reformat feature as the control platform to make the robot control more convenient.
We provide user friendly mobile phone operation interface for remote control by using the blue tooth. The robot has the function of image recognition and transformation, which is the prototype of future robots in human life.
Health
Remote dynamic ECG monitor
PR006 »
Using ADI device and Intel FPGA to measure and process human body ECG signal, upload the signal stream to the cloud for remote diagnoses.
Autonomous Vehicles
Design of Heterogeneous Accelerator for Traffic Sign Classification Algorithm Based on FPGA
PR007 »
This project intends to achieve heterogeneous acceleration of traffic sign classifier algorithms. The recognition of traffic signs is an important part of automatic driving. For traffic sign recognition algorithms, the accuracy of convolutional neural networks is in the forefront, but Existing neural networks and deep neural networks both rely on high-speed parallel computing equipment with high cost and high power consumption. Therefore, it is of practical significance to find a low-cost and alternative to high-cost GPU acceleration solution. FPGA acceleration is one of the choices.
Data Management
拥抱未来
PR010 »
电子工程系参赛队伍
Health
Family health assistant
PR014 »
Integrate the sphygmomanometer, stethoscope and thermometer into one platform and realize the intelligent evaluation of measurement results. At the same time, it will have the function of reminding measurement every day, upload the measurement results to the cloud and form a report. Doctors in the hospital can realize remote consultation by observing the daily measurement data, give reasonable medical suggestions and provide medical help for patients who are inconvenient to move
Smart City
Environmental Humidity Control System Based on RISC-V
PR040 »
Use Intel's FPGA development platform to design an intelligent environment adjustment system driven by a 32-bit RISC-V architecture processor. The system embeds the RISC-V soft core in the FPGA development platform, drives peripheral sensors, and realizes the shortest path optimization algorithm. Finally, it can control the mobile platform in a specific environment and realize the detection and adjustment of environmental humidity with the optimal path. Real-time obstacle avoidance function is realized by distance sensor during movement. In the process of system development, the system will be tested and improved.