IMPLEMENTASI CONVOLUTIONAL NEURAL NETWORK (CNN) DALAM PENGENALAN EKSPRESI WAJAH MANUSIA
Total View This Week0
Total View This Week0
Institusion
Universitas Pembangunan Nasional Veteran Jakarta
Author
Puspita Cahyani Putri, .
Subject
Z719 Libraries (General)
Datestamp
2022-01-13 02:20:05
Abstract :
Face expression recognition is the recognition of emotions based on the expression or feelings of someone resulting from the situation or situation that is being experienced. This study uses Deep Learning with the Convolutional Neural Network (CNN) method which aims to classify human facial expressions and test the accuracy of the CNN method. This research was conducted in several stages, namely conditioning the dataset consisting of 140 images, the process of image improvement (pre-processing), training of image data (training), and testing of image data (testing). The training process is done by comparing the value of epoch, batch size, and learning rate to get high accuracy values. The convolution layer used in this study is four layers followed by a maxpooling layer in each layer. Accuracy results generated from this study were 92% in the training data and 71% in the validation data.
Face expression recognition is the recognition of emotions based on the expression or feelings of someone resulting from the situation or situation that is being experienced. This study uses Deep Learning with the Convolutional Neural Network (CNN) method which aims to classify human facial expressions and test the accuracy of the CNN method. This research was conducted in several stages, namely conditioning the dataset consisting of 140 images, the process of image improvement (pre-processing), training of image data (training), and testing of image data (testing). The training process is done by comparing the value of epoch, batch size, and learning rate to get high accuracy values. The convolution layer used in this study is four layers followed by a maxpooling layer in each layer. Accuracy results generated from this study were 92% in the training data and 71% in the validation data.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
