Using CNN and HOG Classifier to Improve Facial Expression Recognition

Simple item page
dc.contributor.authorOkemwa, Joshua Gisemba
dc.contributor.authorMageto, Victor
dc.date.accessioned2025-03-06T11:59:04Z
dc.date.available2025-03-06T11:59:04Z
dc.date.issued2019-06
dc.descriptionJournal Article
dc.description.abstractFacial expression recognition (FER) is growing on a large scope due to diversification of its field of application. FER is now applicable in crime prevention, smart city development, as well as other economic sectors like: transportation, advertising and health. There are a number of benefits accompanied through proper and correct analysis of emotions: Security is enhanced, proper event prediction, inter person communication channel, easy extraction of details and so on. There are various facial emotions identified from various previous studies, which make up the basis for effective and affective communication among people of different culture, race, ethnicity and gender. Many feature extraction methods and classification techniques have previously been developed to give better accuracy and performance in face recognition. A convolution neural network CNN is an unsupervised deep learning algorithms with ability to learn image characteristics and make differentiation of one aspect to another. It is a trending technique in this field due to its positive results, and fast computation. However, the still there are issues with accuracy and complexity challenges in face recognition. In this paper we perform experiments on FER to solve problems associated with orientation and different light conditions. We applied HOG classifier for feature extraction and CNN to detect and classify the expressions. Overall we achieved high accuracy and optimization results of 77.2%. This method achieved higher results than previous work done using SVM algorithm and HOG classifier with accuracy of 55%.
dc.identifier.citationOkemwa, J. G.& Mageto, V. (2019). Using CNN and HOG Classifier to Improve Facial Expression Recognition. International Journal for Research in Applied Science & Engineering Technology
dc.identifier.issn2321-9653
dc.identifier.urihttps://repository.daystar.ac.ke/handle/123456789/6393
dc.language.isoen
dc.publisherInternational Journal for Research in Applied Science & Engineering Technology
dc.subjectCNN
dc.subjectFER
dc.subjectHOG
dc.subjectSVM
dc.titleUsing CNN and HOG Classifier to Improve Facial Expression Recognition
dc.typeArticle

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Using CNN and HOG Classifier to Improve Facial Expression Recognition.pdf
Size:
906.49 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Journal Articles