Face Recognition from Low Resolution Images: Interpolation of Low Resolution Images for Improved Accuracy in Human Face Recognition

In a wide range of face recognition applications, such as the surveillance camera in law enforcement, it is cannot provide enough resolution of face for recognition. The first part of this research demonstrates the impact of the image resolution on the performance of the face recognition system. The performance of several holistic face recognition algorithms is evaluated for low-resolution face images. For the classification, this research uses k-nearest neighbor (k-NN) and Extreme Learning Machine-based neural network (ELM). The recognition rate of these systems is a function in the image resolution. In the second part of this research, nearest neighbor, bilinear, and bicubic interpolation techniques are applies as a preprocessing step to increase the resolution of the input image to obtain better results. The results show that increasing the image resolution using the mentioned interpolation methods improves the performance of the recognition systems considerably.