Background subtraction plays an important role in intelligent video surveillance since it is one of the most used tools in motion detection. If scientific progress has enabled to develop sophisticated equipment for this task, algorithms used should be improved as well. For the past decade a background subtraction technique called ViBE is gaining the field. However, the original algorithm has two main drawbacks. The first one is ghost phenomenon which appears if the initial frame contains a moving object or in the case of a sudden change in the background situations. Secondly it fails to perform well in complicated background. This paper presents an efficient background subtraction approach based on ViBE to solve these two problems. It is based on an adaptive radius to deal with complex background, on cumulative mean and pixel counting mechanism to quickly eliminate the ghost phenomenon and to adapt to sudden change in the background model.

Recognition performance of biometric systems is affected through spoofing attacks made by fake identities. The focus of this paper is on presenting a new scheme based on score level and decision level fusion to monitor individuals in term of real and fake. The proposed fake detection scheme involve consideration of both handcrafted and deep learned techniques on face images to differentiate real and fake individuals. In this approach, convolutional neural network (CNN) and overlapped histograms of local binary patterns (OVLBP) methods is used to extract facial features of images. The produced matching scores provided by CNN and OVLBP then combined to form a fused score vector. Finally, the last decision on real and attack images is done by combining decisions of hybrid scheme using majority vote of CNN, OVLBP and their fused vector. Experimental results on public spoof databases such as Print-Attack and Replay-Attack face databases demonstrate the strength of the proposed anti-spoofing method for fake detection. 

High returning rate of garments products have become a notable problem for online fashion shopping. This problem is partially caused by using different standards for measuring cloth sizes on different websites. In this research, we have designed a set of equipment to capture images of t-shirts of any color and propose an automatic cloth measurement approach using image processing techniques. A method has been introduced to recognize feature points, which has been used to calculate the cloth sizes. The method has provided a useful and efficient tool for cloth measurement. The photographs have been taken in a controlled environment, and then clothes have been categorized with the proportions of the neck, shoulder, chest width, upper waist, lower waist, and length. In this method, we have measured the t-shirt size for men by calculating the chest width and length of men. For this, a dataset has been created in a specific environment. This method has integrated with a web-based application. We have validated our work by calculating RMSE values.

The Coronary Artery Disease (CAD) which is one among the major class of cardiovascular diseases is emerging as an epidemic in the society and has proven to be the leading cause for more number of deaths when compared to the other cardiovascular diseases. It is emerging as one of the threats to the economy. It has become very important to detect CAD in its early stage which can help society in a broader way by saving a significant number of lives. The proposed method is a novel efficient automated approach which is capable of detecting CAD among the large group of patients using Electrocardiogram (ECG) signal. The system design provides a complete model of pre-processing of ECG, finding the heart rate which is further decomposed up to 4 level sub-bands using analytic transformation based signal decomposition method. The signal decomposition method is used to analyze the low frequency components of the signal and to deal with non stationary nature of heart signals. Two Non-linear entropy estimators as K-Nearest Neighbor (K-NN) and Correlation entropy are applied to decomposed sub- bands obtained after applying Analytic wavelet transformation based flexible decomposition technique to extract non-linear dynamics. The clinical significant features from the large data set can be selected by employing wilcoxon ranking method which assigns ranks on the applied signal. Further, an entropy-based classification approach and a suitable classifier namely Linear support vector machine (L-SVM) is used to classify among CAD and normal class. The algorithm is simulated in MATLAB and it is found that the results matched closely with the available data. This computer-assisted automated system which characterizes the heart signal can serve as an aid for the cardiologists in their daily screening of a large number of patients and can be used in primary health care centers which help the physicians in the early detection of a CAD.  

Control of the quadrotor has been noted for its difficulty as the result of the so-called high maneuverability, exceedingly nonlinear system and strongly coupled multivariable. This work deals with the simulation depend on proposed controllers of a quadrotor that can overcome this difficulty. The quadrotor mathematical model is derived using a Newton-Euler formulation. Three types of controllers are investigated to control and stabilization the position and attitude of quadrotor using feedback linearization. The first controller is Fuzzy-PID, it is considered as a reference benchmark to compare its results with the others two controllers which are PID tuned using GA and ANFIS. The performance of the designed control structure is evaluated through the response and minimizing the error of the position and attitude. Simulation results, shows that position and attitude control using Fuzzy-PID has fast response and better steady state error and RMS error than ANFIS and PID tuned using GA. The all controllers are tested by simulation under the same conditions using SIMULINK under MATLAB2015a.

Arthritis is one of the chronic joint disorders that have affected many lives including middle age and older age group. Arthritis exists in many forms and one among them is Osteoarthritis. Osteoarthritis affects the bigger joints like knee, hip, spine, feet etc. Early detection of Osteoarthritis is most essential if not treated properly may result in deformity. The researchers have become more concerned to detect the disorder in the early stage by merging their medical knowledge with machine vision approach in an appropriate way. The objective of this work is to study various segmentation techniques for the detection of Osteoarthritis in the early stage. The different segmentation technique like Sobel and Prewitt edge segmentation, Otsu’s method of segmentation and Texture based segmentation are used to carry out the experimentation. The different statistical features are computed, analyzed and classified. The accuracy rate of 91.16% for Sobel method, 96.80% for Otsu’s method, 94.92% for texture method and 97.55% for Prewitt method is obtained. The results are more promising and competitive which are validated by medical experts.