Gender Classification and Age Detection Based on Human Facial Features Using Multi- Class SVM

In Human being the key characteristic feature is the face. The various emotions of a human can be determined and easily realized by the different facial expressions. The face is considered to be the most acceptable biometric trait than any other component as image capturing and prediction of images

Project Title

Gender Classification and Age Detection Based on Human Facial Features Using Multi- Class SVM

Project Area of Specialization

Artificial Intelligence

Project Summary

In Human being the key characteristic feature is the face. The various emotions of a human can be determined and easily realized by the different facial expressions. The face is considered to be the most acceptable biometric trait than any other component as image capturing and prediction of images is easier to perform than other traits. Faces are normally classified as semi-rigid, semi-flexible, culturally significant, and part of our individual entity, and thus needs good computing techniques for face recognition and classification.

In this project an attempt has been made to perform gender classification of human facial image based on the extracted feature from the input image set. The feature based on which gender recognition has been performed is the ‘lip’ which has been identified and extracted from a human facial image by following the Region of Interest (ROI) principle. Then the extracted feature is fed as input to the Support Vector Machine Classifier (SVM) which has the data set containing a combined set of lip images of both male and female. Besides this, the above mentioned technique has been implemented on a group image from which individual images of male/ female has been extracted. Then the similar procedure of Feature Extraction and Gender Classification is performed. Finally an attempt has been made to perform age detection of a combined set of images and based on the age the images has been classified into either ‘child’ or ‘adult’ or ‘old’. This classification is performed using Multi- Class SVM.

Project Objectives

To detect the gender of person based on facial features.

To detect the age of person that is adult old or child based on facial features

Project Implementation Method

The model accepts a set of male and female face image in .jpeg format. Each RGB image was then converted to greyscale. Lip feature extraction is performed for each image and is used for classification of each image into either of the two classes (Male and Female) using a SVM classifier.

A set of 100 Jpeg images are selected as the input image set for the experiment of which 50 are of male images and 50 are of female images. Each image is of size 128*115. The input image set are pre-processed ,  subjected to the svmtrain() routine for the generation of the train vector and then test images are applied to the svmclassify() routine where the test data set are tested and the class labels are determined.

FOR AGE DETECTION USING MULTICLASS- SVM

A set of 119 Jpeg images have been considered for the input image set of which ‘child image’ and ‘adult image’ are of equal ratio of 40 images each. The remaining 39 images are the images of old people. Each image has been resized to the dimension of 128*115.

On these image set, histogram equalization is performed which improves the contrast of the input image set. Then we perform Feature Extraction and Age Detection from the input data points.

In the process of Age Detection and classification, the’ “One against All” Strategy of Multi- class SVM has been utilized. Based on the principle of this strategy, three different SVM classifier is created for the three classes of data point’s i.e child, adult and old

Benefits of the Project

Gender recognition has found its strong applications in fields of authentication, search engine accuracy, demographic data collection, human computer interaction, access control and surveillance, involving frontal facial images. It can also be used as indexing technique to reduce the search space for automatic face recognition.

Moreover, there are several other applications where gender recognition plays a crucial role which includes biometric authentication, hightechnology surveillance and security systems, image retrieval, and passive demographical data collections

Technical Details of Final Deliverable

In this project, novel methodologies has been proposed to achieve the goal of (1) gender classification and (2) age detection in three step process. Firstly, input image set are pre- processed to perform noise removal, histogram equalization, size normalization and then face detection is performed. Secondly, Feature Extraction from facial image is performed. Finally to evaluate the performance of the proposed algorithm, experiments have been performed on various image set that contain equal proportion of male and female by using suitable binary SVM classifier which will classify the data set into two categories i.e male or female. To achieve the second goal, Multi- class SVM have been employed which will generate three classes i.e child, adult and old. The age of the input images are detected and classified into one of the three category.

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Education

Core Technology

Artificial Intelligence(AI)

Other Technologies

Internet of Things (IoT)

Sustainable Development Goals

Good Health and Well-Being for People

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com