Detection of Age and Gender from Facial Images Using CNN

This work focuses on the existing application of gender and age classification from an image. In this paper, convolutional neural network is used to accurately classify a person’s gender and age through a single face image. Gender can be predicted to one of ‘Male’ or ‘Female’ and age predicted to be one of the following: (0–2), (4–6), (8–12), (15–20) (25–32), (38–43), (48–53), (60–100) from 8 nodes. Guessing the accurate age of an individual through image can be difficult due to variables such as make-up, lighting, and expressions. Therefore, rather than using regression, we consider it a classification model. Evaluations on the ‘Audience’ dataset, with a combination of simple preprocessing, obtained the performance in gender and age recognition. We experienced some patterns like face recognition, face detection, age recognition by implementing convolutional neural networks (CNNs) for age and gender predictions. The automatic age and gender detection has become increasingly important, particularly as social networks and social media which are rapidly grown. This demonstrates that the use of deep neural networks (CNNs) is increasing significantly to showcase the performance of observed accuracy for age and gender classification which are higher than another existing claim. We used the cloud services to compute and accessing the dataset from cloud storage.