Age Group Estimation Model using K-Nearest Neighborhood
Moka Uma Devi1, Uppu Ravi Babu2
1Moka Uma Devi, Pursuing Ph D, Computer Science & Engineering, Acharya Nagarjuna University, Guntur, India.
2Dr.U.Ravi Babu, Professor, Department of Computer Science and Engineering, Acharya Nagarjuna University, Guntur, India.
Manuscript received on November 25, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 3851-3858 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B4420129219/2019©BEIESP | DOI: 10.35940/ijeat.B4420.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Age estimation labels exact real age or age group for a given face image. How to recognise the face of a human depends upon the age invariant features and patterns. After finding out the aging patterns, the researchers are in investigation to find out in what way we can characterise the aging of a face to get accurate performance. We can estimate the age through multi class classification or regression or a combination of both classification and regression. In our paper we are classifying, predicting and evaluating our proposed aging pattern algorithm to estimate the age. The brief process is first we split the data in to two subsets i.e. training data and test data by using stratified cross validation method. By using training data and test data we are classifying or predicting the age group using K-neighbourhood method and evaluation measures are considered by using confusion matrix. The Classification and Evaluation of Age estimation models results us to find out the best estimation model for different types of datasets which are used in different applications like biometric, law enforcement, and security control and human-computer interaction.
Keywords: Age estimation, K neighbourhood, Multiclass confusion matrix, Prediction, Evaluation