Facial recognition using multi edge detection and distance measure

Indo Intan, Nurdin Nurdin, Fitriaty Pangerang

Abstract


Face recognition provides broad access to several public devices, so it is essential in cutting-edge technology. Human face recognizing has challenge in using uncomplicated and straightforward algorithms quickly, using memory specifications are not too high, otherwise the results are quality and accurate. Face recognition using combination edge detection and Canberra distance can be recommended for applications that require fast and precise access. The application of several edge detections singly has low performance, so it requires a combination technique to obtain better results. The proposed method combined several edge detections such are Robert, Prewitt, Sobel, and Canny to recognize a face image by identification and verification. As a feature extractor, the combination edge detection forms a more robust and more specific facial pattern on the contour lines. The results show that the combination accuracy outperforms other extractor features significantly. Canberra distance produces the best performance compared to Euclidean distance and Mahalanobis distance.

Keywords


Canberra distance; Combination edge detection; Distance measure; Edge detection; Face recognition

Full Text:

PDF


DOI: http://doi.org/10.11591/ijai.v12.i3.pp1330-1342

Refbacks

  • There are currently no refbacks.


View IJAI Stats

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.