In this paper we have developed a robust human identification scheme from low-resolution surveillance video footage. For establishing the human identity; we also carried out a comparative study with two multimodal biometric databases (UCMG and CASIA), two different gait databases with different complexities. For experimental validation of our scheme, we used several dimensionality reduction algorithms (to reduce the dimensionality of the features) and examined number of classifiers to learn the identity model. This study established that gait biometric along with appropriate intelligent processing approaches can allow automatic identity verification from low resolution video surveillance footage. Several physiological and biomechanical studies have shown that human gait is a unique, an inherently multimodal biometric, and involves a complex kinematic interaction between several motion articulators, and includes interplay between lower and upper limbs and other biomechanics of joints(1). In this paper, we propose that usage of full profile silhouettes (full gait) of persons from multi-view low resolution cameras for capturing inherently multi-modal cues available from the gait patterns of the walking humans, and use it for establishing their identity. Further, we propose the use of unsupervised feature learning techniques, based on variants of principal component analysis (vPCA) and deep learning (DL) approaches, which allow an in- depth analysis of the underlying pixel data. We compare these new features to standard features based on multivariate statistical techniques, such as PCA and linear discriminant analysis (LDA), along with well-known learning classifier approaches based on support vector machines, NN and MLP classifiers(2) (3), (4), (5). The experimental evaluation of the proposed approach with two different databases, the publicly available CASIA (6) gait database, and the newly developed UCMG database (7), show a significant improvement in recognition performance with the proposed unsupervised learning features, particularly for uncooperative camera conditions, simulated with mismatched train and test data sets.
This article is available (Open Access) to view at: http://www.globalcis.org/dl/citation.html?id=JNIT-320
In bioinformatics, one of the goldstandard algorithms to compute the optimal similarity score between sequences in a sequence database searches is Smith-Waterman algorithm that uses dynamic programming. This algorithm has a quadratic time complexity which requires a long computation time for large-sized data. In this issue, parallel computing is essential for sequence database searches in order to reduce the running time and to increase the performance. In this paper, we discuss the parallel implementation of Smith-Waterman algorithm in GPU using CUDA C programming language with NVCC compiler on Linux environment. Furthermore, we run the performance analysis using three parallelization models, including Inter-task Parallelization, Intra-task Parallelization, and a combination of both models. Based on the simulation results, a combination of both models has better performance than the others. In addition the parallelization using combination of both models achieves an average speed-up of 313x and an average efficiency with a factor of 0.93.
In this paper, the virtual flight simulation environment for open-source-based avionics test system design and implementation. This paper is proposing an integrated avionics test system for flight manipulation and simulation using the commercial simulation and development software tools such as X-Plane, LabView, and Google Earth. The proposed system is designed of seven parts of user flight control, flight simulation, flight manipulation and verification, data link and distribution, map display, data interface, and actuation systems. For a specific planned flight path, it is found that the proposed system was well operated and could be used in the flight test system.
Tae-Sub Kim, Seung-Yeon Kim, Seungwan Ryu, Hyong-Woo Lee, Choong-Ho Cho 1 First Author Dept.of Computer and Information Science, Korea University, Korea 2,4 Dept.of Information Systems, Chung-Ang University, Korea 3 Dept.of Electronics and Information Engineering, Korea University, Korea *5 Corresponding Author Dept.of Computer and Information Science, Korea University, Korea {ree31206, kimsy8011, hwlee, chcho}@korea.ac.kr, ryu@cau.ac.kr
Abstract Ant colony system which is classified as a meta-heuristic algorithm is considered as one of the best optimization algorithm for solving different type of NP-Hard problem including the travelling salesman problem. A heuristic function in the Ant colony system uses pheromone and distance values to produce heuristic values in solving the travelling salesman problem. However, the heuristic values are not updated in the entire process to reflect the knowledge discovered by ants while moving from city to city. This paper presents the work on enhancing the heuristic function in ant colony system in order to reflect the new information discovered by the ants. Experimental results showed that enhanced algorithm provides better results than classical ant colony system in term of best, average and standard of the best tour length. Keywords : Ant Colony Optimization, Ant Colony System, Heuristic Function, Traveling Salesman Problem 1. Introduction Biological ants have the ability to discover the shortest route from the nest to the source of food [1]. Although they do not have an advanced vision system [2], they have the ability to communicate with the environment. Ants use a chemical substance called a “pheromone” to communicate with the environment and between each other [3]. Pheromone substance has an evaporation property which is a powerful mechanism to update the route information. While an ant moves looking for food, it deposits a pheromone along the path. The following ant will, more likely, select the route with richer pheromones. This mechanism will make the ant choose the shortest path. In 1992, Marco Dorigo proposed the first Ant Colony Optimization (ACO) algorithm to search for an optimal solution in graphs to solve optimization problems such as the travelling salesman problem, job scheduling and network routing [1]. The variants of ACO are: (i) Ant System (AS) [4] [5] [6]. (ii) The first improvement on the ant system, called the Elitist strategy for Ant System (EAS) [7]. The improvement was done by providing strong additional reinforcement to the arcs belonging to the best tour found since the start of the algorithm. (iii) Rank-Based Ant System (AS
Vehicle routing is one of the crucial logistic activities for delivering goods or services. This paper demonstrates the application of the clonal selection of Artificial Immune System (AIS), Generalized Evolutionary Walk Algorithm (GEWA) and Genetic Algorithm (GA) for solving capacitated vehicle routing problem (CVRP). The optimal parameter settings of the proposed algorithms were investigated using statistical design and analysis of experiment. Sequentially, the performances of the proposed algorithms were compared using twenty benchmarking CVRP instant datasets. The results showed that the parameters of each algorithm were statistically significant with a 95% confidence interval. It was found that the best-so-far solutions obtained from the AIS were up to 71.4% more efficient than those produced by the GEWA and GA for all problem sizes but the AIS required longer computational time.
Information retrieval involves finding most relevant information within a large amount of documents and provides various forms of contents for people who search information. The user information need should be presented as a query. Finding relevant texts through the user's query is difficult by growing texts in internet. A model which is used to find the similarities between user's query and available information is the first step of designing information retrieval systems. Several models were presented during decades. In this paper, we carefully study information retrieval models and discuss about advantages and disadvantages of each model. We also classify these models into three main classes based on mathematical rules: set theoretic model, algebraic model and probabilistic model. Each class has some sub classes which are carefully studied . Finally we propose some metrics to evaluate these models and compare them with each other.
This paper proposes an active capacitive sensing circuit for fingerprint sensors, which includes a pixel level charge-sharing and charge pump to replace an ADC. This paper also proposes the operating algorithm for 16-level gray scale image. The active capacitive technology is more flexible and can be adjusted to adapt to a wide range of different skin types and environments. The proposed novel circuit is composed with unit gain buffer, 6-stage charge pump and analog comparator. The proper operation is validated by the HSPICE simulation of one pixel with condition of 0.35μm typical CMOS parameter and 3.3V power.
During procurement, enterprises tend to choose the bid with the lowest price as preferred choice; thus, suppliers tend to win the bid with low prices and then deliver low-quality goods. If the enterprises fail to detect the low quality goods, they may suffer from financial losses. In order to solve this problem, a system for determining the most advantageous bid has been proposed, which sets the criterion and sub-criterion for supplier selection, as well as the weights of criterion and sub-criterion. By literature review and expert interviews, this study determined the criteria and sub-criteria for the selection of the most advantageous bid, and conducted a questionnaire survey on procurement experts. According to expert opinions, this study used the Fuzzy Analytic Hierarchy Process (FAHP) to calculate the weights of criterion and sub-criterion of the most advantageous bid. Findings of this study can serve as reference for enterprises selecting the most advantageous bid.
This paper develops a methodology in how to apply customer journey methods in designing technology-mediated services. The focus is particularly on how to improve customer satisfaction and perceived value through introducing innovative services delivered through self-service technologies (SSTs). Action research is applied on a sub-program of a four-year-long government sponsored project in Taiwan. A methodology covering nine steps is proposed, which is believed valuable in bridging perspectives of service marketing, innovation design, and IT-enabled service (ITeS) development. The strategic roles and implications of customer journey methods in identifying current service gaps and diagnosing the potentials of SST-mediated services are also spotlighted.
People reasons to choose an application is for the functionalities offered. Though, the way in which the functions are performed will also have a significant impact to the user’s choice. Software usability is one of the crucial factors that can influence user acceptance. It assesses the extent of which software facilitates users in utilizing the offered functions easily and effectively. Usability requirements are associated and complement the functional requirements. However, it is common that the usability requirements are captured at the design stage of software development due to its characteristic. Essentially, it should be identified at the earlier stage (i.e., requirements stage) to ensure the success of the software product. Therefore, we explore the potential benefits of Soft System Methodology (SSM) in identifying usability requirements at the requirements stage. The proposed SSM-based framework is applied in preliminary analysis, in which the results are useful to inform the next stage of the development process. The practical benefits of the proposed analysis framework are demonstrated in a case study. The framework would be beneficial particularly for mobile software developers in improving the quality of the mobile applications.
Under the shallow sea condition, bottom reverberation is the main background noise. Especially the bottom and the buried targets echo are weak and signal-to-reverberation ratio is very low. So it is an important problem that how to improve the signal-to-reverberation ratio. This paper presents a method which is Perturbation Iteration Method. The method can reduce the intensity of bottom reverberation and improve the signal-to-reverberation ratio without changing the measurement environment and measurement equipments. This method sums the received signals to offset a portion of the unrelated components of the reverberation, using the different correlation radius between bottom reverberation and the target signal’s. This method’s physical meaning is clear and so simple to understand. At the same time it can avoid the complex signal processing methods. The simulation results show that the method can reduce the intensity of the bottom reverberation effectively, and can detect the target under low signal-to-reverberation cases.
The aim of this study was to design and evaluate a Digital Game Based Learning software for history subject. The methodology used was Instructional Design-Digital Game Based Learning (IDDGBL) Model adapted from ADDIE and game development as well as 5E intructional models. Design elements include learning theories, pedagogy and game components. The effectiveness evaluation used quasi-experimental tecnique via pre and post tests. The samples involved were 60 secondary school students; 30 in experimental group who used the software after class lesson and 30 in control group who used conventional method. Both the independent t test result show that there is significant difference of achievements between the control group (mean score = 35.80) and the experimental group (mean score = 72.30), p 4.00) design for all constructs evaluated (multimedia, interface, contents, feedback, ease of use and immersion). Motivation and learning opporturnity for history DGBL software are also high. Evaluation results indicate that the History DGBL software can entertain, attract, engage and motivate students to learn history subject. Students had fun and felt entertained learning through illustration of historical events when they play the games immersively. Evaluation results suggest that history DGBL software can be used as an alternative teaching and learning tool to promote student’s learning of
Diabetes is one of the modern chronic diseases that have a big impact on the life of a large number of populations across the world. According to the World Health Organization, it is estimated that this disease causes an annual death of up to 5% globally. Therefore there would be an urgent need to create system that is capable of aiding healthcare providers to monitor and then manage such chronic disease. Such a system requires a mechanism that helps the diagnosing of the patients and to participate in the aid of an eHealth management system. The aim of this research is to create such a mechanism upon the usage of Artificial Neural Networks that is capable of predicting the condition of a diabetic patient based on certain number of factors that are associated with such a chronic disease. The learning process was carried out using training cases that were extracted from local diabetic patients’ records. Specialized diabetes physicians were consulted in order to create an accurate training model. The latter model was then tested and results were evaluated and presented as part of this paper.
