The main problem in face recognition system based on half-face pattern is how to anticipate poses and illuminance variations to improve recognition rate. To solve this problem, we can use two lenses on stereo vision camera in face recognition system. Stereo vision camera has left and right lenses that can be used to produce a 2D image of each lens. Stereo vision camera in face recognition has capability to produce two of 2D face images with a different angle. Both angle of the face image will produce a detailed image of the face and better lighting levels on each of the left and right lenses. In this study, we proposed a face recognition technique, using 2 lens on a stereo vision camera namely symmetrical half-join. Symmetrical half-join is a method of normalizing the image of the face detection on each of the left and right lenses in stereo vision camera, then cropping and merging at each image. Tests on face recognition rate based on the variety of poses and variations in illumination shows that the symmetrical half-join method is able to provide a high accuracy of face recognition and can anticipate variations in given pose and illumination variations. The proposed model is able to produce 86% -97% recognition rate on a variety of poses and variations in angles between 0 °- 22.5 °. The variation of illuminance measured using a lux meter can result in 90% -100% recognition rate for the category of at least dim lighting levels (above 10 lux).

/pdf/face-recognition-based-on-symmetrical-half-join-method-using-525km696ge.pdf

Face Recognition Based on Symmetrical Half-Join Method using Stereo Vision Camera

This research proposes a model of face recognition using the method of joining two face images from left and right lens from a stereo vision camera namely half-join method. The half-join method is used in face image normalization processing. The proposed half-join method is a face image joining model, which is called asymmetrical half-join. In the asymmetrical half-join method, a RoI (region of interest) of face image from left and right lenses are provided based on axis center of each eye in eye detection. The cropping of face image from the asymmetrical half-join model has different width depends on eyes coordinate location. The proposed system shows that the asymmetrical half-join method can produce a better of face recognition rate. The experimental results show that the asymmetrical half-join method has a better recognition rate and computation time than single vision method and symmetrical half-join method.

/pdf/asymmetrical-half-join-method-on-dual-vision-face-45kh4xagat.pdf

Asymmetrical Half-join Method on Dual Vision Face Recognition

Visual Speech Recognition (VSR) is to identify spoken words from visual data only without the corresponding acoustic signals. It is useful in situations in which conventional audio processing is ineffective like very noisy environments or impossible like unavailability of audio signals. In this paper, an optimized model for VSR is introduced which proposes simple geometric projection method for mouth localization that reduces the computation time.16-point distance method and chain code method are used to extract the visual features and its recognition performance is compared using the classifier Hidden Markov Model (HMM). To optimize the model, more prominent features are selected from a large set of extracted visual attributes using Discrete Cosine Transform (DCT). The experiments were conducted on an in-house database of 10 digits [1 to 10] taken from 10 subjects and tested with 10-fold cross validation technique. Also, the model is evaluated based on the metrics specificity, sensitivity and accuracy. Unlike other models in the literature, the proposed method is more robust to subject variations with high sensitivity and specificity for the digits 1 to 10. The result shows that the combination of 16-point distance method and DCT gives better results than only 16-point distance method and chain code method.

An optimized model for visual speech recognition using HMM.

In this paper, a new Face Recognition method based on Two Dimensional Discrete Cosine Transform with Linear Discriminant Analysis (LDA) and K Nearest neighbours (KNN) classifier is proposed. This method consists of three steps, i) Transformation of images from special to frequency domain using Two dimensional discrete cosine transform ii) Feature extraction using Linear Discriminant Analysis and iii) classification using K Nearest Neighbour  classifier. Linear Disceminant Analysis searches the directions for maximum discrimination of classes in addition to dimensionality reduction. Combination of Two Dimensional   Discrete Cosine transform and Linear Discriminant Analysis is used for improving the capability of Linear Discriminant Analysis when few samples of images are available. K Nearest Neighbour classifier gives fast and accurate classification of face images that makes this method useful in online applications. Evaluation was performed on two face data bases. First database of 400 face images from AT&T face database, and the second database of thirteen students are taken. The proposed method gives fast and better recognition rate when compared to other classifiers. The main advantage of this method is its high speed processing capability and low  computational requirements in terms of both speed and memory utilizations. DOI: http://dx.doi.org/10.11591/ij-ai.v1i4.767

https://ijai.iaescore.com/index.php/IJAI/article/download/1295/1186

I. V. Murali Krishna

Papers

Face Recognition Using Two Dimensional Discrete Cosine Transform, Linear Discriminant Analysis And K Nearest Neighbor Classifier