Prediction of the Age and Gender Based on Human Face Images Based on Deep Learning Algorithm

Area of Science:

• Computer Science
• Artificial Intelligence
• Nutrition Science

Background:

• Personalized nutrition advice is crucial for healthy living, yet current systems often lack individual-specific recommendations.
• Existing food recommendation systems primarily focus on general health issues rather than tailored nutritional guidance.

Purpose of the Study:

• To develop an intelligent nutrition recommendation system that provides personalized food suggestions based on age and gender.
• To leverage facial image analysis for accurate age and gender prediction to drive nutritional recommendations.

Main Methods:

• Image preprocessing and feature extraction using Deep Convolution Neural Network (DCNN).
• Feature selection employing Hybrid Particle Swarm Optimization (HPSO) to identify distinctive facial traits.
• Age and gender classification using Support Vector Machine (SVM).

Main Results:

• The system accurately predicts age and gender from facial images.
• Demonstrated excellent performance in classification rate, precision, and recall on Adience and UTKface datasets.
• Achieved efficient computation time for real-world image analysis.

Conclusions:

• The proposed system effectively integrates facial recognition with nutrition recommendations for personalized health.
• Facial image analysis offers a viable approach for developing individualized nutrition guidance systems.
• The system shows significant potential for promoting healthy lifestyles through tailored dietary advice.