▍1. 卷积神经网络识别调制信号星座图
利用tensorflow构建了一个三层的卷积神经网络识别星座图的调制方式。数据源由matlab生成,存储为301*301的图片格式。在训练50次左右准确率可以达到99%以上
利用tensorflow构建了一个三层的卷积神经网络识别星座图的调制方式。数据源由matlab生成,存储为301*301的图片格式。在训练50次左右准确率可以达到99%以上
说明: DNN检测入侵网络的用到的数据集是KDDCUP99可以参考(The data set used in DNN intrusion detection network is KDDCUP99, which can be referred to.)
利用tensorflow框架来实现卷积神经网络(Using tensorflow framework to realize convolution neural network)
说明: 利用tensorflow框架来实现卷积神经网络(Using tensorflow framework to realize convolution neural network)
说明: 基于LSTM的雷达高分辨距离像识别算法(anaconda)(Radar high resolution range profile recognition algorithm based on LSTM)
当今,自动驾驶技术已经成为整个汽车产业的最新发展方向。应用自动驾驶技术可以全面提升汽车驾驶的安全性、舒适性,满足更高层次的市场需求等。自动驾驶技术得益于人工智能技术的应用及推广,在环境感知、精准定位、决策与规划、控制与执行、高精地图与车联网 V2X 等方面实现了全面提升。科研院校、汽车制造厂商、科技公司、自动驾驶汽车创业公司以及汽车零部件供应商在自动驾驶技术领域进行不断地探索,寻求通过人工智能技术来获得技术上的新突破。本报告在此背景下,对自动驾驶汽车进行了简单梳理(Nowadays, automatic driving technology has become the latest development direction of the whole automobile industry. The application of automatic driving technology can comprehensively improve the safety and comfort of automobile driving, and meet the higher level of market demand. Auto-driving technology benefits from the application and promotion of artificial intelligence technology, and has achieved comprehensive improvement in environmental awareness, precise positioning, decision-making and planning, control and execution, high-precision map and V2X. Research institutes, automobile manufacturers, technology companies, autopilot auto startups and auto parts suppliers continue to explore in the field of automatic driving technology, seeking to achieve new technological breakthroughs through AI technology. Under this background, this report makes a simple combing of autopilot cars.)
本算法实现了InceptionV3模型的迁移学习。训练好的inceptionV3模型可自行搜索下载.pb文件,数据集需为本地jpg图片。(Realization of full adder schematic diagram)