Home / All Innovations / Innovation Details

Safety Driving based on fast and high accuracy Face and Eyes Detection Algorithm v1.0

Description: The aim of the project is to achieve safe driving by detecting the driver's face and eye position, and we expect this project to be easily installed on the vehicle. At the same time, the system hardware needs to use a computer combined with LifeCam to achieve. The system software will test several different datasets to test human face and eye position. The human face part tests five datasets, which contains CMU face dataset, FERER face dataset, MIT face dataset, ORL face dataset and YALE face dataset. The eye position part tests three datasets including MRL eyes dataset, CEW eyes dataset and EGG eyes dataset. Find the best accuracy data to establish training model. The execution time of the system is expected to be close to real-time, so we will use several NN algorithms for testing. Among the many Neural Network algorithms, we use PCA combined with Support Vector Machine(SVM)、Extreme Learning Machine (ELM),Kernel - Extreme Learning Machine(K-ELM) and Sparse Bayesian - Extreme Learning

Organisation: HouKong Middle School


Category: Information Technology

Country: Macau

Silver Award