Facial emotion recognition finds many real applications in the daily life like human robot interaction, eLearning, healthcare, customer services etc. The task of facial emotion recognition is not easy due to the difficulty in determining the effective feature set that can recognize the emotion conveyed within the facial expression accurately. Graph mining techniques are exploited in this paper to solve facial emotion recognition problem. After determining positions of facial landmarks in face region, twelve different graphs are constructed using four facial components to serve as a source for sub-graphs mining stage using gSpan algorithm. In each group, the discriminative set of sub-graphs are selected and fed to Deep Belief Network (DBN) for classification purpose. The results obtained from the different groups are then fused using Naïve Bayes classifier to make the final decision regards the emotion class. Different tests were performed using Surrey Audio-Visual Expressed Emotion (SAVEE) database and the achieved results showed that the system gives the desired accuracy (100%) when fusion decisions of the facial groups. The achieved result outperforms state-of-the-art results on the same database.
This paper presents the application of a framework of fast and efficient compressive sampling based on the concept of random sampling of sparse Audio signal. It provides four important features. (i) It is universal with a variety of sparse signals. (ii) The number of measurements required for exact reconstruction is nearly optimal and much less then the sampling frequency and below the Nyquist frequency. (iii) It has very low complexity and fast computation. (iv) It is developed on the provable mathematical model from which we are able to quantify trade-offs among streaming capability, computation/memory requirement and quality of reconstruction of the audio signal. Compressed sensing CS is an attractive compression scheme due to its uni
... Show MoreThe aim of the research is to estimate the hidden population. Here، the number of drug users in Baghdad was calculated for the male age group (15-60) years old ، based on the Bayesian models. These models are used to treat some of the bias in the Killworth method Accredited in many countries of the world.
Four models were used: random degree، Barrier effects، Transmission bias، the first model being random، an extension of the Killworth model، adding random effects such as variance and uncertainty Through the size of the personal network، and when expanded by adding the fact that the respondents have different tendencies، the mixture of non-random variables with random to produce
... Show MoreQuantum dots of CdSe, CdS and ZnS QDs were prepared by chemical reaction and used to fabricate organic quantum dot hybrid junction device. QD-LEDs were fabricated using layers of ITO/TPD: PMMA/CdSe/Alq3, ITO/TPD: PMMA/CdS/Alq3 and ITO/TPD: PMMA/ZnS/Alq3 devices which prepared by phase segregation method. The hybrid white light emitting devices consists, of three-layers deposited successively on the ITO glass substrate; the first layer was of N, N’-bis (3-methylphenyl)-N, N’-bis (phenyl) benzidine (TPD) polymer mixed with polymethyl methacrylate (PMMA) polymers. The second layer was QDs while the third layer was tris (8-hydroxyquinoline) aluminium (Alq3
... Show MoreWe report on using a CO2 (10.6 µm) laser to debond the lithium disilicate veneers. Sixty-four sound human premolar teeth and 64 veneer specimens were used in the study. The zigzag movement via CO2 laser handpiece along with an air-cooled jet to prevent temperature elevation above the necrosis temperature limit (5.5 C°) was applied. The optimal deboning irradiation time was super-fast, at about 5 seconds at 3 Watt CO2 laser power. It is 20 times less than any previously published work for veneers debonding. The enamel beneath the debonded veneers has been assessed by atomic force microscopy (AFM) and shear stress technique as criteria for the easiness of debonding. The
... Show MoreAbstract
The methods of the Principal Components and Partial Least Squares can be regard very important methods in the regression analysis, whe
... Show MoreSeeds of five cultivars of oats (Avena sativa) were introduced from Italy in 2009. Seeds were propagated on the farm of the Dept. of Field Crops Sci. / Coll. of Agric. / Univ. of Baghdad in the season 2009 – 2010. The cultivars Anatolia, Alguda, Hamel, Pimula and Genzania were planted under 3 irrigation intervals; 3, 4 and 5 weeks to give water depth of 480, 400 and 320 mm, respectively . The depth of water was 80 mm each irrigation. A factorial experiment with RCBD of 4 replicates was conducted in 2 consecutive seasons in 2010 – 2011 and 2011 – 2012. The cultivar Alguda gave highest grain yield (8.07 t/ ha) under 480 mm, and 7.02 t / ha average of 3 water depths. This cultivar was characterized by high growth rate (13.2 g/m2/ d) that
... Show MoreBackground: Complete analysis of facial profile should also include an evaluation of soft tissue morphology. Materials and Method:The sample consisted of 90 Iraqi adults (45 males and 45 females) aged 18-25 years from Baghdad city divided into 3 groups according to the ANB angle with 30 subjects in each group (15 males and 15 females) for class I, II and III. Lateral cephalometric radiograph was taken for each subject and 8 angular and 5 linear measurements were identified and determined, t-test, ANOVA and LSD test were used to compare between both genders and between different classes. Results:Showed that females had greater angular measurements and smaller linear measurements with more lip prominence than males in all classes, there was m
... Show MoreIn data mining, classification is a form of data analysis that can be used to extract models describing important data classes. Two of the well known algorithms used in data mining classification are Backpropagation Neural Network (BNN) and Naïve Bayesian (NB). This paper investigates the performance of these two classification methods using the Car Evaluation dataset. Two models were built for both algorithms and the results were compared. Our experimental results indicated that the BNN classifier yield higher accuracy as compared to the NB classifier but it is less efficient because it is time-consuming and difficult to analyze due to its black-box implementation.