The process of accurate localization of the basic components of human faces (i.e., eyebrows, eyes, nose, mouth, etc.) from images is an important step in face processing techniques like face tracking, facial expression recognition or face recognition. However, it is a challenging task due to the variations in scale, orientation, pose, facial expressions, partial occlusions and lighting conditions. In the current paper, a scheme includes the method of three-hierarchal stages for facial components extraction is presented; it works regardless of illumination variance. Adaptive linear contrast enhancement methods like gamma correction and contrast stretching are used to simulate the variance in light condition among images. As testing material

<span>One of the main difficulties facing the certified documents documentary archiving system is checking the stamps system, but, that stamps may be contains complex background and surrounded by unwanted data. Therefore, the main objective of this paper is to isolate background and to remove noise that may be surrounded stamp. Our proposed method comprises of four phases, firstly, we apply k-means algorithm for clustering stamp image into a number of clusters and merged them using ISODATA algorithm. Secondly, we compute mean and standard deviation for each remaining cluster to isolate background cluster from stamp cluster. Thirdly, a region growing algorithm is applied to segment the image and then choosing the connected regi

A multidimensional systolic arrays realization of LMS algorithm by a method of mapping regular algorithm onto processor array, are designed. They are based on appropriately selected 1-D systolic array filter that depends on the inner product sum systolic implementation. Various arrays may be derived that exhibit a regular arrangement of the cells (processors) and local interconnection pattern, which are important for VLSI implementation. It reduces latency time and increases the throughput rate in comparison to classical 1-D systolic arrays. The 3-D multilayered array consists of 2-D layers, which are connected with each other only by edges. Such arrays for LMS-based adaptive (FIR) filter may be opposed the fundamental requirements of fa

      In this work, a weighted H lder function that approximates a Jacobi polynomial which solves the second order singular Sturm-Liouville equation is discussed. This is generally equivalent to the Jacobean translations and the moduli of smoothness. This paper aims to focus on improving methods of approximation and finding the upper and lower estimates for the degree of approximation in weighted H lder spaces by modifying the modulus of continuity and smoothness. Moreover, some properties for the moduli of smoothness with direct and inverse results are considered.

A three-stage learning algorithm for deep multilayer perceptron (DMLP) with effective weight initialisation based on sparse auto-encoder is proposed in this paper, which aims to overcome difficulties in training deep neural networks with limited training data in high-dimensional feature space. At the first stage, unsupervised learning is adopted using sparse auto-encoder to obtain the initial weights of the feature extraction layers of the DMLP. At the second stage, error back-propagation is used to train the DMLP by fixing the weights obtained at the first stage for its feature extraction layers. At the third stage, all the weights of the DMLP obtained at the second stage are refined by error back-propagation. Network structures an

This paper deals with calculate stresses in Knee-Ankle-Foot-Orthosis as a result of the effect vibration during gait cycle for patient wearing KAFO .Experimental part included measurement interface pressure between KAFO and leg due to action muscles and body weigh on Orthosis. also measurement acceleration result from motion of defected leg by accelerometer .Results of Experimental part used input in theoretical part so as to calculate stresses result from applying pressure and acceleration on KAFO by engineering analysis program ANSYS 14.Resultes show stresses values in upper KAFO greater than lower KAFO that is back to muscles more effective in thigh part lead to recoding pressure higher than pressure in shank part.

This research examines the use of vibratory treatments to reduce residual stresses in small welded parts. In this experimental investigation, a post weld vibration treatment was applied to T- A106 steel pipe fitting specimens to study the effect of the treatment on the residual stress and the hardness of the material. The vibratory stress relief treatment was carried out at different vibration frequency. The results have demonstrated that post-weld vibratory stress relief of small size fittings is possible and residual stress may be relieved, and the treatment may be an alternative method for heat treatment especially when unchange in dimensions and material stability are required.