STATEMENT OF THE PROBLEM

Nowadays, computerization is one of the prior advancement of technology in everyday life. Everyday activities of people depend themselves to technology just to make their work easy and fast. In every company always do attendance before they get into work because it serves that they are present. One of the problems on this transaction is automation of attendance, they are using the old manual process or identification card sheet just to have list of attendance. On this process, it can affect the work of the people. Companies should propose and produce a kind of system that help them and make their attendance process easy, fast and accurate with the use of advancement of the technology. Through this process or system, it can easily recognize their selves so they can log in and out fast, accurately, easier and less time effort.

FOREIGN RELATED LITERATURE In this scenario, the eyeglass companys use of facial detection to simply locate the consumers face in an image the consumer has voluntarily uploaded does not, by itself, raise privacy concerns. However, there may be privacy concerns related to the companys collection and storage of the consumers image in connection with its virtual fitting feature. The eyeglass company should take several steps to address these concerns. First, it should design its service with privacy in mind by implementing privacy by design. For instance, Commission staff agrees with panelists who noted that companies that hold databases of consumer images should take steps to protect those images.43 In this example, the eyeglass company should have reasonable data security protections in place for any stored images in order to prevent unauthorized access to those images. In addition, the eyeglass company should implement a specified retention period and dispose of stored images once they are no longer necessary for the purpose for which they were collected. If a consumer deletes his or her account on the website, the stored images

are no longer necessary and should be disposed of, even if the stated retention period has not yet passed.
Second, consistent with panelist and commenter recommendations on transparency and choice, the company should be clear with consumers about its data practices and provide choices appropriate for the context of the transaction. One appropriate way to achieve transparency in this scenario is to state at the time the consumer uploads her image to try on glasses why the company is storing, rather than immediately deleting, that image (e.g., Create an account so you can try on glasses in the future). If the company is storing the images for a purpose that is not consistent with the context of the transaction taking place, it should provide additional information about why it is storing the images at a just in time point. For example, if the company stores the images for purposes of sharing them with

third parties, it should explicitly provide consumers with a choice about this practice before they upload their image outside of a privacy policy or similar document. In all cases, the company should also inform consumers of: the length of time the images are stored, who will have access to the stored images, and Consumers rights regarding deletion of the stored images.

Finally, consistent with the recommendations in the Commissions Privacy Report, if in the future the eyeglass company decides to use the images in a materially different manner than it represented at the time of collection, the company should obtain the affirmative express consent of the consumer prior to such use. For instance, if the eyeglass company decides to use the images in its advertising, rather than simply storing them for future use by the consumer, this would require the consumers affirmative express consent.

------------------------------------------------------------------------------------------------------------------------------------------

Face recognition has been an active research area over last 30 years. This research spans several disciplines such as image processing, pattern recognition, computer vision, and neural networks. It has been studied by scientists from different areas of psychophysical sciences and those from different areas of computer sciences. Psychologists and neuroscientists mainly deal with the human perception part of the topic, whereas engineers studying on machine recognition of human faces deal with the

computational aspects of face recognition. Face recognition has applications mainly in the fields of biometrics, access control, law enforcement, and security and surveillance systems. The problem of face recognition can be stated as follows: Given still images or video of a scene, identifying one or more persons in the scene by using a stored database of faces [23]. The problem is mainly a classification problem. Training the face recognition system with images from the known individuals and classifying the newly coming test images into one of the classes is the main aspect of the face recognition systems. This problem seems to be easily solved by humans where limited memory can be the main problem; whereas the problems for a machine face recognition system are: 1. Facial expression change 2. Illumination change 3. Aging 4. Pose change 5. Scaling factor (i.e. size of the image) 6. Frontal vs. pro_le 7. Presence and absence of spectacles, beard, mustache etc. 8. Occlusion due to scarf, mask or obstacles in front. The problem of automatic face recognition (AFR) is a composite task that involves detection of faces from a cluttered background, facial feature extraction, and face Identification. A complete face recognition system has to solve all sub problems, where each one is a separate research problem. This research work concentrates on the problem of facial feature extraction and face identification. Most of the current face recognition algorithms can be categorized into two classes, image template based and

geometry feature-based. The template based methods [9] compute the correlation between a face and one or more model templates to estimate the face identity. Brunelli and Poggio [16] suggest that the optimal strategy for face recognition is holistic and corresponds to template matching. In their study, they compared a geometric feature based technique with a template matching based system and reported an accuracy of 90% for the first one and 100% for the second one on a database of 97 persons. Statistical tools such as Support Vector Machines (SVM) [92, 127], Principal Component Analysis (PCA) [114, 124], Linear Discriminant Analysis (LDA) [12], kernel methods [109, 136], and neural networks [50, 74, 97] have been used to construct a suitable set of face templates. Other than statistical analysis and neural network approach there are other approaches known as hybrid approaches which use both statistical pattern recognition techniques and neural network systems. Examples for hybrid approaches include the combination of PCA and Radial Basis Function (RBF) neural network [37, 122]. Among other methods, people have used 10 range [23], infrared scanned [137] and pro_le [79] images for face recognition. While templates can be viewed as features, they mostly capture global features of the face image. Facial occlusion is often difficult to handle in these approaches. The geometry feature based methods analyze explicit local facial features, and their geometric relationships. Cootes et al. [72] have presented an active shape model in extending the approach by Yuille [139]. Wiskott et al. [131] developed an elastic bunch graph matching algorithm for face recognition. Penev et al. [95] developed PCA into Local Feature Analysis (LFA). This technique is the basis for one of the most successful commercial face recognition systems, FaceIt. The summary of approaches to face recognition is shown in Fig. 2.1.

-----------------------------------------------------------------------------------------------------------------

A flurry of research has emerged providing solutions to various heterogeneous face recognition problems. This began with sketch recognition using viewed sketches, and has continued into other modalities such as near-infrared (NIR) and forensic sketches. In this section we will highlight a representative selection of studies in heterogeneous face recognition as well as studies that use kernel based approaches for classification. Tang et al. spearheaded the work in heterogeneous face recognition with several approaches to synthesize a sketch from a photograph (or vice-versa) [82, 138, 149]. Tang and Wang initially proposed an eigen-tranformation method [138]. Later, Liu et al. performed the transformation using local linear embedding to estimate the corresponding photo patch from a sketch patch [82]. Wang and Tang proposed a Markov random eld model for converting a sketch into a photograph [149]. Other synthesis methods have been proposed as well [29, 157]. A key advantage of synthesis methods is that once a sketch has been converted to a photograph, matching can be performed using existing face recognition algorithms. The proposed prototype framework is similar in spirit to these methods in that no direct comparison between face images in the probe and gallery modalities is needed. The generative transformation-based approaches have generally been surpassed by discriminative feature-based approaches. A number of discriminative feature-based approaches to HFR have been pro- posed [12, 52, 59, 77], which have shown good matching accuracies in both the sketch and NIR domains. These approaches rst represent face images using local feature descriptors, such as variants of local binary patterns (LBP) [99] and SIFT descriptors [84]. Liao et al. first used this approach on NIR to VIS face recognition by processing face images with a di erence of Gaussian lter, and encoding them using multi-block local binary patterns (MB-LBP). Gentle AdaBoost feature selection was used in conjunction with R-LDA to improve the recognition accuracy. Klare and Jain followed this work on NIR to VIS face recognition by also incorporating SIFT feature descriptors and an RS-LDA scheme [52]. Bhatt et al. introduced an extended uniform circular local binary pattern to the viewed sketch recognition scenario [12]. Klare et al. encoded both viewed sketches and forensic sketches using SIFT

and MLBP feature descriptors, and performed local feature-based discriminant analysis (LFDA) to improve the recognition accuracy [59]. Yi et al. [154] local patch-based method to perform HFR on partial NIR face images. The synthesis method by Li et al. is the only known method to perform recognition between thermal IR and visible face images [71]. The only method to perform recognition between forensic sketches and visible face images is Klare et al. [59], which is also one of two methods, to our knowledge, that has been tested on two different HFR scenarios (viewed sketch and forensic sketch). The other method is Lin and Tang's [78] common discriminant recognition framework which was applied to viewed sketches and near-infrared images. In this work the proposed prototype random subspace framework is tested on four different HFR scenarios.