www.ijecs.in International Journal Of Engineering And Computer Science Volume 9 Issue 06 June 2020, Page No. 25070-25074 ISSN: 2319-7242 DOI: 10.18535/ijecs/v9i06.4500 Rahul S Hangal, IJECS Volume 09 Issue 06 June, 2020 Page No. 25070-25074 Page 25070 Identical Twins Facial Recognition System Using Cloud 1 Chandrakala G Raju, 2 Rahul S Hangal * , 3 Shashidhara A R, 4 Srinatha T D 1 Assistant Professor, Department of Information Science and Engineering, B.M.S College of Engineering, Basavangudi, Bengaluru, Karnataka, India 2,3,4 Department of Information Science and Engineering, B.M.S College of Engineering, Basavangudi, Bengaluru, Karnataka, India Abstract: Facial recognition algorithm should be able to work even when the similar looking people are found i.e. also in the extreme case of identical looking twins. An experimental data set which contains 40 images of 20 pairs of twins collected randomly from the internet. The training is done with the selected images of the twins using different training algorithms and inbuilt functions available. The extracted features are stored over the Amazon public cloud. As a part of testing phase random images from the dataset trained are selected and upon running it over the system we get the features of those images which then will be compared by extracting the features already stored in Amazon cloud. The stored values and the current image features are compared and result will be displayed on the GUI. Identical twin’s facial recognition system uses the machine learning, image processing algorithms and deep learning algorithms. Regardless of the conditions of the images acquired, distinguishing identical twins is significantly harder than distinguishing faces that are not identical twins for all the algorithms. Keywords: Identical Twins, Cloud, Facial Recognition, Image Processing. 1. Introduction In computer technology classifying the images based on identical twin’s facial recognition is a difficult task. Old facial recognition system shows very poor performance in finding difference in identical twins. Traditionally the experiments were performed to find out the difference between identical twins and also to recognize their features with difference and system like finger print, voice and iris recognition show the difference between the identical twins. In the existing methods many technologies are used for twin’s identification like the finger print, the voice and the iris for the recognition. The finger print identification was used to identify the unique person. The method also proposed a scan image taken from the person and that image being compared to that in the database for identification. The iris recognition also has a similar method to that of the finger print identification system. This proposed system in this system will differentiate the identical looking twins using face recognition system. 2. Existing Systems The existing systems are as follows: 1. Facial Recognition System 2. Eyes or Iris Recognition System 3. Fingerprint Recognition System 4. Voice Recognition System 2.1 Facial Recognition System Facial Recognition system and biometric identification is done by scanning the human face and there by matching the with the stored images in the database. Face recognition system can be used as a platform to identify and verify a person. 2.2 Eyes or Iris Recognition System Eyes are the important feature that is available on the face of a person. The structure of the eye may vary for everyone. Iris biometrics system will exploit the textural information from the iris that have been shown to be independent even between the irises of genetically identical faces of the