Project

Deep learning with convolutional neural networks for image recognition: step-by-step process from preparation to generalization

Project (M.S., Computer Science)--California State University, Sacramento, 2018.

This project collects several experiments in Deep Learning Convolutional Neural Network for Image predictions. It makes use of Google TensorFlow and TFlearn Deep Learning libraries for computations, training, and testing of images. The project is developed in Python language on Linux Operating System. It makes use of TensorFlow on CPU and has the capability to implement on GPU as well. The scope of the project is defined for 4 different sets of data to show how Convolutional Neural Network is architecture independent. The final step is to prepare my own dataset which can be trained and tested for Facial Recognition.
 
 The final experiment walks through the entire process on a custom dataset – from image preparation, through neural network configuration, training, and then testing for generalization.

This project collects several experiments in Deep Learning Convolutional Neural Network for Image predictions. It makes use of Google TensorFlow and TFlearn Deep Learning libraries for computations, training, and testing of images. The project is developed in Python language on Linux Operating System. It makes use of TensorFlow on CPU and has the capability to implement on GPU as well. The scope of the project is defined for 4 different sets of data to show how Convolutional Neural Network is architecture independent. The final step is to prepare my own dataset which can be trained and tested for Facial Recognition. The final experiment walks through the entire process on a custom dataset – from image preparation, through neural network configuration, training, and then testing for generalization.

Relationships

Items