Open AccessAge Prediction using Image Dataset using Machine Learning
Author(s) -
Ishita Verma,
Urvi Marhatta,
Sachin Sharma,
Vijay Kumar
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.l1020.10812s319
Subject(s) - computer science , artificial intelligence , convolutional neural network , face (sociological concept) , preprocessor , feature (linguistics) , artificial neural network , machine learning , pattern recognition (psychology) , facial recognition system , feature extraction , social science , linguistics , philosophy , sociology
Gender is a central feature of our personality still. Inour social life it is also an significant element. Artificialintelligence age predictions can be used in many fields, such assmart human-machine interface growth , health, cosmetics,electronic commerce etc. The prediction of people's sex and agefrom their facial images is an ongoing and active problem ofresearch. The researchers suggested a number of methods toresolve this problem, but the criteria and actual performance arestill inadequate. A statistical pattern recognition approach forsolving this problem is proposed in this project.ConvolutionaryNeural Network (ConvNet / CNN), a Deep Learning algorithm, isused as an extractor of features in the proposed solution. CNNtakes input images and assigns value to different aspects / objects(learnable weights and biases) of the image and can differentiatebetween them. ConvNet requires much less preprocessing thanother classification algorithms. While the filters are hand-madein primitive methods, ConvNets can learn these filters / featureswith adequate training.In this research, face images ofindividuals have been trained with convolutionary neuralnetworks, and age and sex with a high rate of success have beenpredicted. More than 20,000 images are containing age, genderand ethnicity annotations. The images cover a wide range ofposes, facial expression, lighting, occlusion, and resolution.