PENGARUH INTENSITAS CAHAYA TERHADAP HASIL PENGENALAN CITRA DENGAN GRAY LEVEL CO-OCCURRENCE MATRIX DAN PROBABILISTIC NEURAL NETWORK
Face recognition is a basic method of developing an authentication system using the natural characteristics of the human face as baseline. This facial image recognition process through the training phase of the training face images with MATLAB programs and test phases were performed directly on the face images are sourced directly from the camera and not on test data derived from a set of face images that have been selected. Introduction The method combines Canny, GLCM and PNN, the pretreatment stage that converts the RGB image into gray level and the Canny edge detection to determine the reference point as a separator edge areas that are not used. Face recognition factor tested was taken on the influence of the intensity of illumination of the object. GLCM are used with energy parameters, correlation, homogeneity and contrast. PNN while comparing the output of data from the GLCM matrix results. This study uses a database of facial images with a sample of 20 people in the 7 position of the face, distance 3, and 5 categories lighting. The results obtained with the database in a light that category 4 and a distance of 30 cm obtained the highest level of recognition in the process of introducing the light category 5 and 30 cm distance with the level of recognition accuracy of 82.86 percent.
This work is licensed under aÂ Creative Commons Attribution 4.0 International License.