Detector the face (face detection) was one of the beginning stages that was very important in the introductory system to the face (face recognition) that was used in the identification biometrik. The detection of the face could be also used for the search or pengindeksan the face data from the image or the video that contained the face with various measurements, the position, and the background. In this research diimplementasikan the detection system of the face that used the network of nerves of the imitation of the multi-layer perceptron and the Quickprop algorithm.
The detection system of the face received his detection capacity by means of learning from the example (learning by examples). The training was carried out with the method active learning to minimise the number of datas that was used for the training. Results of the research showed that the accuracy from the detection system of the face really depended on the number and the data kind that were used in the training. The Quickprop algorithm and the technique active learning could be made use of to help sped up the process of the training.
The detection system of the face received his detection capacity by means of learning from the example (learning by examples). The training was carried out with the method active learning to minimise the number of datas that was used for the training. Results of the research showed that the accuracy from the detection system of the face really depended on the number and the data kind that were used in the training. The Quickprop algorithm and the technique active learning could be made use of to help sped up the process of the training.
0 comments:
Post a Comment