Facial expression detection from low resolution facial images with deep learning method

Facial expressions play an important role in communicating with people. Apart from communication, existing problems are solved by inferences made from facial expressions in many fields such as security systems, the health sector, and psychology. With the development of technology, inferences to be made from facial expressions are now made thanks to artificial intelligence. Detection of micro mimics is of great importance in detecting facial expressions, and artificial intelligence achieves great success at this point. 213 low-resolution images used in this study were enhanced and images were analyzed using convolutional neural networks (CNN), which is one of the deep learning models. AlexNet and MobileNetV2 architectures, which are deep learning architectures, were used and these features were combined by extracting features from both architectures. The patch-based method was used to detect micro mimics. Separate features were extracted from the image itself and from the patches. The first 1000 features were selected with the NCA (neighborhood component analysis) algorithm and the classification process was carried out with the SVM (support vector machine). The obtained results were compared and the best result was calculated as 98.6%.