• Raveendra K
• Ravi J

The success rate of face recognition is influenced by illumination, expression, posture change and other variables thus use of Convolutional Neural Network (CNN) alone has a weak generalization capability. A novel technique to facial recognition based on coupled learning of convolutional neural networks and Local Phase Quantization (LPQ) is proposed as a solution to this problem. First, face texture features are extracted using the LPQ operator. Once the layer is fully connected, two convolutional neural networks are utilized to extract more training data, improve network parameters, and deliver classification results using the softmax function. Using this method, the recognition rates in the NIR and Yale face datasets are enhanced to 90% and 93.333%, respectively. The results of the studies demonstrate that the suggested method not only increases the model's tolerance for illumination, expression, and posture but also increases the accuracy of face recognition.

"A novel technique to facial recognition based on coupled learning of convolutional neural networks and local phase quantization ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 7, page no.g174-g183, July-2022, Available :http://www.jetir.org/papers/JETIR2207625.pdf

"A novel technique to facial recognition based on coupled learning of convolutional neural networks and local phase quantization ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 7, page no. ppg174-g183, July-2022, Available at : http://www.jetir.org/papers/JETIR2207625.pdf