International Journal of Engineering Research and General Science

6-17

Authors: Ibtisam Naveed1, Adnan Sabir2

The induction motors were characterized by complex, highly non-linear and time-varying dynamics, and hence their speed control is a challenging problem in the industry. The aim of this paper is to present the speed control of three phase induction motors by variable frequency drive. The variable frequency drive for induction motors is achieved using SPWM which provide better efficiency and higher performance. The motor speed is investigated both at fixed load and variable load. Moreover a close loop PI controller was designed at rated load based on dynamic behavior of error signal. It has been found that by designing a proper PI controller the motor starting current is reduced significantly. Moreover at rated torque efficiently speed of motor can be controlled. PI speed controller not help to reduce dynamic performance of the system but also help to reduce the steady state error, the error sensibility, high performance and smooth speed response. The complete mathematical model of the system is described and simulated in MATLAB/SIMULINK. The simulation results provide a smooth speed response and high performance under various dynamic operations.Keywords— Induction Motor, PI controller, SPWM technique, Voltage Source Inverter (VSI)

24-31

Authors: Laveena Agarwal1, Vinish Kumar2, Dwaipayan Dey 3

From last few years, vehicle plate recognition (VPR) has become a vital technology of security and traffic applications. In this paper we present a training based approach for the recognition of vehicle number plate. The whole process has been divided into three stages i.e. capturing the image, plate localization and recognition of digits over the plate. HOG features have been used for the training purpose and Support Vector Machine (SVM) is employed for the classification purpose yielding in more than 99% accuracy while recognition. The algorithm has been tested over more than 100 images. Keywords: Vehicle Plate Recognition (VPR), Support Vector Machine (SVM), Histogram of Gradient (HOG),Optical Character Recognition (OCR), Licence Plate Recognition (LPR)

38-47

Authors: Ali M. Hasan

The Mid-Sagittal Plain (MSP) of Brain is an important initial step in brain image analysis because of providing an initial estimation for the brain, pathology assessment and tumor detection. The human brain is divided into two hemispheres and these hemispheres are approximately bilateral symmetry around the MSP, that’s mean that most of structures in one side of the brain have a counterpart on the other side with similar shape and location. This paper presents a technique for detecting and correcting the MSP of brain from magnetic resonance images (MRI) by using Principle component analysis (PCA) method that computes the distinctive principle axes that are orthogonal to each other, and determine the orientation of the patient’s head instead of depending on measuring symmetry to identify the MSP of brain. Keywords-component; Magnetic resonance image (MRI); Principle compnent analysis (PCA); Mid-Sagittal Plain (MSP)

75-82

Authors: Anuja C.G1, Anand J Dhas2

This paper highlights a non-invasive 3D reconstruction and monitoring technique for early detection of melanoma. Melanoma is life threating when it grows beyond the dermis of skin. Hence, depth is an important factor to diagnose melanoma. The computer aided monitoring system empowers and motivates the user to actively manage their skin health status by collecting, processing and store information of skin lesion through classifications. In this paper introduces a 3D segmentation method, which is an energy minimization based method designed to segment individual objects in 3D satisfying certain size and shape properties. Different feature set combination is considered and performance is evaluated. Experimental results prove that the proposed computerized dermoscopic system is efficient and can be used to diagnose varied skin lesions. Keywords— 3D lesion reconstruction, 3D features and tumor depth estimation, 3D segmentation, skin lesions, melanoma, Bag of Features, Adaboost algorithm, Support Vector Machine.

91-100

Authors: Abin O. Bright 1, J.Blessy2

Automatic aircraft recognition is a challenging task. Conventional methods always extract the overall shapes of aircraft at first and then represent the aircraft based on the extracted shape with different features for recognition. The major problem of these methods is that they have a high requirement on shape extraction, which is too idealistic for targets in Remote sensing images. To address these challenges, propose a new aircraft recognition framework, which can be divided into three processes: region proposal, classification, and accurate aircraft localization process. First, a region proposal method is used to generate candidate regions with the aim of detecting all objects of interest within these images. Then, generic image features from a local image corresponding to each region proposal are extracted by a combination model of 2-D reduction convolutional neural networks (CNNs). Haar cascade method is used to classification of aircraft. To improve the location accuracy, introduces an unsupervised score based bounding box regression (USB-BBR) algorithm, combined with a non-maximum suppression algorithm to optimize the bounding boxes of regions that detected as objects. Experiments show that the dimension-reduction model performs better than the retrained and fine-tuned models and the detection precision of the combined CNN model is much higher than that of any single model. The proposed USB-BBR algorithm can more accurately locate objects within an image. Compared with traditional features extraction methods, such as elliptic Fourier transform based histogram of oriented gradients and local binary pattern histogram Fourier, the proposed localization framework shows robustness when dealing with different complex backgrounds. Keywords— Aircraft recognition, Convolutional neural networks (CNN), unsupervised score-based bounding box regression (USB-BBR), Haar Cascade method.

109-114

Authors: Nkodi* T.M1, Kayembe J.S2., Biey M. E 3.and Taba K.M2

This work aims to study co-digestion of composted mixture of cassava peels and coffee pulp with cow dung .Cassava peels was mixed with coffee pulp in the following ratios by weight, 1:1, 3:1, 1:3, Cassava peels alone (1:0) and coffee pulp alone (0:1) served as controls. The mixtures and controls were submitted to composting for 30 days then placed in five digesters C1 for 1:1 blend,C2 for 3:1 blend,C3 for 1:3 blend , C4 for cassava peels alone and C5 for coffee pulps alone. Another batch of five digesters consisting of each composted mixture of cassava peels and coffee pulps to which was added equal amount of cow dung labeled as follows: CB1, CB2, CB3, CB4 and CB5. Biodigestion was conducted under mesophilic conditions into 5L biodigesters with a loading rate of 5% of total solids. Biogas production was measured by using water displacement method for 35 days. The highest cumulative volume of biogas was produced by composted mixture with cow dung and the yield decreased in the following order 16.50, 15.38, 15.07, 12.68 and 10.97L/KgTS respectively for CB3>CB1>CB5>CB2>CB4. The composted mixture alone yielded the following amount of biogas: 10.26, 9.11, 8.86, 7.50, and 7.07L/KgTS respectively for C1, C5, C4, C3 and C2. Co-digestion of the composted mixture of cassava peels and coffee pulps with cow dung can be considered as a good system to enhance biogas production from cassava peels.