Figure 4 - uploaded by Dua'A Ali Kareem
Content may be subject to copyright.
Similar publications
The adaptability of novometric analysis is illustrated for an example involving three class categories and two ordered attributes.
Citations
... The preprocessing phases in this paper consist of two stages that seek to enhance the visual appearance of the image [18]. The first stage, which is image transform to grey level by transforming the three components of the color image RGB bands into one band, will be noted as grey as shown in (1) 595 ...
This paper presents a simple and fast recognition system with various facial expressions, poses, and rotation. The proposed system performed in two phases. Face detection is the first phase. The front and profile face detected cropped face area from the image by Viola-Jones algorithm and the right side face is detected from the image by taking the flip of the profile image. Principal component analysis (eigenfaces) algorithm is used in the recognition phase and depends on created database models used to be compared with test face image input to the recognition procedure. For training and testing the system, two sets of the image of the file exchange interface (FEI) database have been used to identify the person. The experimental result shows the effectiveness and robustness of the method used for the detection of the face and achieves high accuracy of 96%, which improves the recognition performance with low execution time. Furthermore, the accuracy of 35 trained images for recognition is 97.143% with average time execution which is (0.323657s). Also, the accuracy of 15 tested images for recognition is 93.315% with average time execution which is (0.3348s) which indicates a good and strong success and accurate method for facial recognition.
Recently, one of the most important biometrics is that automatically recognized human faces are based on dynamic facial images with different rotations and backgrounds. This paper presents a real-time system for human face tracking and recognition with various expressions of the face, poses, and rotations in an uncontrolled environment (dynamic background). Many steps are achieved in this paper to enhance, detect, and recognize the faces from the image frame taken by web-camera. The system has three steps: the first is to detect the face, Viola-Jones algorithm is used to achieve this purpose for frontal and profile face detection. In the second step, the color space algorithm is used to track the detected face from the previous step. The third step, principal component analysis (eigenfaces) algorithm is used to recognize faces. The result shows the effectiveness and robustness depending on the training and testing results. The real-time system result is compared with the results of the previous papers and gives a success, effectiveness, and robustness recognition rate of 91.12% with a low execution time. However, the execution time is not fixed due depending on the frame background and specification of the web camera and computer.</span
