Abstract: As a rapid growth of digital videos and data
communications, video summarization that provides a shorter version
of the video for fast video browsing and retrieval is necessary.
Key frame extraction is one of the mechanisms to generate video
summary. In general, the extracted key frames should both represent
the entire video content and contain minimum redundancy. However,
most of the existing approaches heuristically select key frames; hence,
the selected key frames may not be the most different frames and/or
not cover the entire content of a video. In this paper, we propose
a method of video summarization which provides the reasonable
objective functions for selecting key frames. In particular, we apply
a statistical dependency measure called quadratic mutual informaion
as our objective functions for maximizing the coverage of the
entire video content as well as minimizing the redundancy among
selected key frames. The proposed key frame extraction algorithm
finds key frames as an optimization problem. Through experiments,
we demonstrate the success of the proposed video summarization
approach that produces video summary with better coverage of
the entire video content while less redundancy among key frames
comparing to the state-of-the-art approaches.
Abstract: In this paper, we propose a framework to help users to search and retrieve the portions in the lecture video of their interest. This is achieved by temporally segmenting and indexing the lecture video using the topic keywords. We use transcribed text from the video and documents relevant to the video topic extracted from the web for this purpose. The keywords for indexing are found by applying the non-negative matrix factorization (NMF) topic modeling techniques on the web documents. Our proposed technique first creates indices on the transcribed documents using the topic keywords, and these are mapped to the video to find the start and end time of the portions of the video for a particular topic. This time information is stored in the index table along with the topic keyword which is used to retrieve the specific portions of the video for the query provided by the users.
Abstract: Clustering is a well known data mining technique used in pattern recognition and information retrieval. The initial dataset to be clustered can either contain categorical or numeric data. Each type of data has its own specific clustering algorithm. In this context, two algorithms are proposed: the k-means for clustering numeric datasets and the k-modes for categorical datasets. The main encountered problem in data mining applications is clustering categorical dataset so relevant in the datasets. One main issue to achieve the clustering process on categorical values is to transform the categorical attributes into numeric measures and directly apply the k-means algorithm instead the k-modes. In this paper, it is proposed to experiment an approach based on the previous issue by transforming the categorical values into numeric ones using the relative frequency of each modality in the attributes. The proposed approach is compared with a previously method based on transforming the categorical datasets into binary values. The scalability and accuracy of the two methods are experimented. The obtained results show that our proposed method outperforms the binary method in all cases.
Abstract: Choosing good features is an essential part of machine learning. Recent techniques aim to automate this process. For instance, feature learning intends to learn the transformation of raw data into a useful representation to machine learning tasks. In automatic audio classification tasks, this is interesting since the audio, usually complex information, needs to be transformed into a computationally convenient input to process. Another technique tries to generate features by searching a feature space. Genetic algorithms, for instance, have being used to generate audio features by combining or modifying them. We find this approach particularly interesting and, despite the undeniable advances of feature learning approaches, we wanted to take a step forward in the use of genetic algorithms to find audio features, combining them with more conventional methods, like PCA, and inserting search control mechanisms, such as constraints over a confusion matrix. This work presents the results obtained on particular audio classification problems.
Abstract: Achieving an appropriate assembly sequence needs deep verification for its physical feasibility. For this purpose, industrial engineers use several assembly predicates; namely, liaison, geometric feasibility, stability and mechanical feasibility. However, testing an assembly sequence for these predicates requires huge assembly information. Extracting such assembly information from an assembled product is a time consuming and highly skillful task with complex reasoning methods. In this paper, computer aided methods are proposed to extract all the necessary assembly information from computer aided design (CAD) environment in order to perform the assembly sequence planning efficiently. These methods use preliminary capabilities of three-dimensional solid modelling and assembly modelling methods used in CAD software considering equilibrium laws of physical bodies.
Abstract: In this paper, we mainly analyze an automated parking system where the storage and retrieval requests are performed by a tower crane. In this parking system, the S/R crane which is located at the middle of the bottom of the cylinder parking area can rotate in both clockwise and counterclockwise and three kinds of movements can be done simultaneously. We develop some mathematical travel time models for the single command cycle under the random storage assignment using the characteristics of this system. Finally, we compare these travel models with discrete case and it is shown that these travel models display a good satisfactory performance.
Abstract: This paper focuses on developing a system to transfer mold base plates and standard parts faster during the stage of injection mold design. This system not only provides a way to compare the file version, but also it utilizes Siemens NX 10 to isolate the updated information into a single executable file (.dll), and then, the file can be transferred without the need of transferring the whole file. By this way, the system can help the user to download only necessary mold base plates and standard parts, and those parts downloaded are only the updated portions.
Abstract: Image Multi-label Classification (IMC) assigns a label or a set of labels to an image. The big demand for image annotation and archiving in the web attracts the researchers to develop many algorithms for this application domain. The existing techniques for IMC have two drawbacks: The description of the elementary characteristics from the image and the correlation between labels are not taken into account. In this paper, we present an algorithm (MIML-HOGLPP), which simultaneously handles these limitations. The algorithm uses the histogram of gradients as feature descriptor. It applies the Label Priority Power-set as multi-label transformation to solve the problem of label correlation. The experiment shows that the results of MIML-HOGLPP are better in terms of some of the evaluation metrics comparing with the two existing techniques.
Abstract: The current state-of-the-art methods of mass gauging of Electric Propulsion (EP) propellants in microgravity conditions rely on external measurements that are taken at the surface of the tank. The tanks are operated under a constant thermal duty cycle to store the propellant within a pre-defined temperature and pressure range. We demonstrate using computational fluid dynamics (CFD) simulations that the heat-transfer within the pressurized propellant generates temperature and density anisotropies. This challenges the standard mass gauging methods that rely on the use of time changing skin-temperatures and pressures. We observe that the domes of the tanks are prone to be overheated, and that a long time after the heaters of the thermal cycle are switched off, the system reaches a quasi-equilibrium state with a more uniform density. We propose a new gauging method, which we call the Improved PVT method, based on universal physics and thermodynamics principles, existing TRL-9 technology and telemetry data. This method only uses as inputs the temperature and pressure readings of sensors externally attached to the tank. These sensors can operate during the nominal thermal duty cycle. The improved PVT method shows little sensitivity to the pressure sensor drifts which are critical towards the end-of-life of the missions, as well as little sensitivity to systematic temperature errors. The retrieval method has been validated experimentally with CO2 in gas and fluid state in a chamber that operates up to 82 bar within a nominal thermal cycle of 38 °C to 42 °C. The mass gauging error is shown to be lower than 1% the mass at the beginning of life, assuming an initial tank load at 100 bar. In particular, for a pressure of about 70 bar, just below the critical pressure of CO2, the error of the mass gauging in gas phase goes down to 0.1% and for 77 bar, just above the critical point, the error of the mass gauging of the liquid phase is 0.6% of initial tank load. This gauging method improves by a factor of 8 the accuracy of the standard PVT retrievals using look-up tables with tabulated data from the National Institute of Standards and Technology.
Abstract: In this paper, a shot boundary detection method is presented using octagon square search pattern. The color, edge, motion and texture features of each frame are extracted and used in shot boundary detection. The motion feature is extracted using octagon square search pattern. Then, the transition detection method is capable of detecting the shot or non-shot boundaries in the video using the feature weight values. Experimental results are evaluated in TRECVID video test set containing various types of shot transition with lighting effects, object and camera movement within the shots. Further, this paper compares the experimental results of the proposed method with existing methods. It shows that the proposed method outperforms the state-of-art methods for shot boundary detection.
Abstract: Nowadays, ontology is common in many areas like artificial intelligence, bioinformatics, e-commerce, education and many more. Ontology is one of the focus areas in the field of Information Retrieval. The purpose of an ontology is to describe a conceptual representation of concepts and their relationships within a particular domain. In other words, ontology provides a common vocabulary for anyone who needs to share information in the domain. There are several ontology domains in various fields including engineering and non-engineering knowledge. However, there are only a few available ontology for engineering knowledge. Fuzzy logic as engineering knowledge is still not available as ontology domain. In general, fuzzy logic requires step-by-step guidelines and instructions of lab experiments. In this study, we presented domain ontology for Fuzzy Logic Control (FLC) knowledge. We give Table of Content (ToC) with middle strategy based on the Uschold and King method to develop FLC ontology. The proposed framework is developed using Protégé as the ontology tool. The Protégé’s ontology reasoner, known as the Pellet reasoner is then used to validate the presented framework. The presented framework offers better performance based on consistency and classification parameter index. In general, this ontology can provide a platform to anyone who needs to understand FLC knowledge.
Abstract: Clustering is an intensive research for some years
because of its multifaceted applications, such as biology, information
retrieval, medicine, business and so on. The expectation maximization
(EM) is a kind of algorithm framework in clustering methods, one
of the ten algorithms of machine learning. Traditionally, optimization
of objective function has been the standard approach in EM. Hence,
research has investigated the utility of evolutionary computing and
related techniques in the regard. Chemical Reaction Optimization
(CRO) is a recently established method. So the property embedded
in CRO is used to solve optimization problems. This paper presents
an algorithm framework (EM-CRO) with modified CRO operators
based on EM cluster problems. The hybrid algorithm is mainly
to solve the problem of initial value sensitivity of the objective
function optimization clustering algorithm. Our experiments mainly
take the EM classic algorithm:k-means and fuzzy k-means as an
example, through the CRO algorithm to optimize its initial value, get
K-means-CRO and FKM-CRO algorithm. The experimental results
of them show that there is improved efficiency for solving objective
function optimization clustering problems.
Abstract: With the emergence and development of Information
and Communications Technologies (ICTs), Higher Education is
experiencing rapid changes, not only in its teaching strategies but
also in student’s learning skills. However, we have noticed that
students often have difficulty when seeking innovative, useful, and
interesting learning resources for their work. This is due to the
lack of supervision in the selection of good query tools. This paper
presents AINA, an Information Retrieval (IR) computer system aimed
at providing motivating and stimulating content to both students
and teachers working on different areas and at different educational
levels. In particular, our proposal consists of an open virtual resource
environment oriented to the vast universe of Disney comics and
cartoons. Our test suite includes Disney’s long and shorts films,
and we have performed some activities based on the Just In Time
Teaching (JiTT) methodology. More specifically, it has been tested
by groups of university and secondary school students.
Abstract: Selecting an appropriate image representation is the most important factor in implementing an effective Content-Based Image Retrieval (CBIR) system. This paper presents a multi-feature fusion approach for efficient CBIR, based on the distance distribution of features and relative feature weights at the time of query processing. It is a simple yet effective approach, which is free from the effect of features' dimensions, ranges, internal feature normalization and the distance measure. This approach can easily be adopted in any feature combination to improve retrieval quality. The proposed approach is empirically evaluated using two benchmark datasets for image classification (a subset of the Corel dataset and Oliva and Torralba) and compared with existing approaches. The performance of the proposed approach is confirmed with the significantly improved performance in comparison with the independently evaluated baseline of the previously proposed feature fusion approaches.
Abstract: Current production-oriented factories need maintenance operators to work in shifts monitoring and inspecting complex systems and different equipment in the situation of mechanical breakdown. Augmented reality (AR) is an emerging technology that embeds data into the environment for situation awareness to help maintenance operators make decisions and solve problems. An application was designed to identify the problem of steam generators and inspection centrifugal pumps. The objective of this research was to find the best medium of AR and type of problem solving strategies among analogy, focal object method and mean-ends analysis. Two scenarios of inspecting leakage were temperature and vibration. Two experiments were used in usability evaluation and future innovation, which included decision-making process and problem-solving strategy. This study found that maintenance operators prefer build-in magnifier to zoom the components (55.6%), 3D exploded view to track the problem parts (50%), and line chart to find the alter data or information (61.1%). There is a significant difference in the use of analogy (44.4%), focal objects (38.9%) and mean-ends strategy (16.7%). The marked differences between maintainers and operators are of the application of a problem solving strategy. However, future work should explore multimedia information retrieval which supports maintenance operators for decision-making.
Abstract: The rich Islamic resources related to religious text,
Islamic sciences, and history are widely available in print and in
electronic format online. However, most of these works are only
available in Arabic language. In this research, an attempt is made
to utilize these resources to create interactive web applications in
Arabic, English and other languages. The system utilizes the Pattern
Recognition, Knowledge Management, Data Mining, Information
Retrieval and Management, Indexing, storage and data-analysis
techniques to parse, store, convert and manage the information from
authentic Arabic resources. These interactive web Apps provide
smart multi-lingual search, tree based search, on-demand information
matching and linking. In this paper, we provide details of application
architecture, design, implementation and technologies employed. We
also presented the summary of web applications already developed.
We have also included some screen shots from the corresponding web
sites. These web applications provide an Innovative On-line Learning
Systems (eLearning and computer based education).
Abstract: When using Information Retrieval Systems (IRS), users often present search queries made of ad-hoc keywords. It is then up to the IRS to obtain a precise representation of the user’s information need and the context of the information. This paper investigates optimization of IRS to individual information needs in order of relevance. The study addressed development of algorithms that optimize the ranking of documents retrieved from IRS. This study discusses and describes a Document Ranking Optimization (DROPT) algorithm for information retrieval (IR) in an Internet-based or designated databases environment. Conversely, as the volume of information available online and in designated databases is growing continuously, ranking algorithms can play a major role in the context of search results. In this paper, a DROPT technique for documents retrieved from a corpus is developed with respect to document index keywords and the query vectors. This is based on calculating the weight (
Abstract: This paper presents and compares the SSDDD “Systematic Soft Domain Driven Design Framework” to DDD “Domain Driven Design Framework” as a soft system approach of information systems development. The framework use SSM as a guiding methodology within which we have embedded a sequence of design tasks based on the UML leading to the implementation of a software system using the Naked Objects framework. This framework has been used in action research projects that have involved the investigation and modelling of business processes using object-oriented domain models and the implementation of software systems based on those domain models. Within this framework, Soft Systems Methodology (SSM) is used as a guiding methodology to explore the problem situation and to develop the domain model using UML for the given business domain. The framework is proposed and evaluated in our previous works, a comparison between SSDDD and DDD is presented in this paper, to show how SSDDD improved DDD as an approach to modelling and implementing business domain perspectives for Information Systems Development. The comparison process, the results, and the improvements are presented in the following sections of this paper.
Abstract: Wireless Sensor Networks (WSNs) are suitable for many scenarios in the real world. The retrieval of data is made efficient by the data aggregation techniques. Many techniques for the data aggregation are offered and most of the existing schemes are not energy efficient and secure. However, the existing techniques use the traditional clustering approach where there is a delay during the packet transmission since there is no proper scheduling. The presented system uses the Velocity Energy-efficient and Link-aware Cluster-Tree (VELCT) scheme in which there is a Data Collection Tree (DCT) which improves the lifetime of the network. The VELCT scheme and the construction of DCT reduce the delay and traffic. The network lifetime can be increased by avoiding the frequent change in cluster topology. Secure and Efficient Transmission of Aggregated data (SETA) improves the security of the data transmission via the trust value of the nodes prior the aggregation of data. Since SETA considers the data only from the trustworthy nodes for aggregation, it is more secure in transmitting the data thereby improving the accuracy of aggregated data.
Abstract: The paper presents the results of DIAL measurements of the vertical ozone distribution. The ozone lidar operate as part of the measurement complex at Siberian Lidar Station (SLS) of V.E. Zuev Institute of Atmospheric Optics SB RAS, Tomsk (56.5ºN; 85.0ºE) and designed for study of the vertical ozone distribution in the upper troposphere–lower stratosphere. Most suitable wavelengths for measurements of ozone profiles are selected. We present an algorithm for retrieval of vertical distribution of ozone with temperature and aerosol correction during DIAL lidar sounding of the atmosphere. The temperature correction of ozone absorption coefficients is introduced in the software to reduce the retrieval errors. Results of lidar measurement at wavelengths of 299 and 341 nm agree with model estimates, which point to acceptable accuracy of ozone sounding in the 6–18 km altitude range.