A decision delay can translate into significant financial and business losses. One way to accelerate the decision process is through improved communication among the stakeholders engaged in the project. Capturing, transferring, managing, and reusing data, information, and knowledge in the context it is generated can lead to higher productivity, effective communication, reduced number of requests for clarification, and a shorter time-to-market cycle. We formalized the concept of reflection in interaction during communicative events among multiple project stakeholders. This concept extends Donald (...) Schon’s theory of reflection in action of a single practitioner. We model the observed reflection in interaction with a prototype system called TalkingPaperTM. It is a ubiquitous client-server collaborative environment that facilitates knowledge capture, sharing, and reuse during synchronous and asynchronous communicative events. TalkingPaperTM bridges the paper and digital worlds. It transforms the analog verbal discours, annotated paper corporate documents, and the paper and pencil sketches into indexed and synchronized digital content that is published on and streamed on-demand from a TalkingPaperTM web server. The TalkingPaperTM sessions can be accessed by all stakeholders for rapid knowledge sharing and decision-making. (shrink)

. Textline segmentation in handwritten documents is of real challenge and interesting in the field of document image processing. In this paper, we propose a handwritten textline segmentation scheme based on the concept of linked list. The proposed method consists of three stages, namely, preprocessing, linked list and mathematical morphology. The concept of the linked list approach is used to build a textline sequence and mathematical morphology is used to obtain the line separator. We experimentally evaluated our proposed method on (...) a document containing handwritten Kannada script. The results are compared with recent methods and show encouraging results. (shrink)

. This paper illustrates the applications of various ensemble methods for enhanced classification accuracy. The case in point is the Pima Indian Diabetic Dataset. The computational model comprises of two stages. In the first stage, k-means clustering is employed to identify and eliminate wrongly classified instances. In the second stage, a fine tuning in the classification was effected. To do this, ensemble methods such as AdaBoost, bagging, dagging, stacking, decorate, rotation forest, random subspace, MultiBoost and grading were invoked along with (...) five chosen base classifiers, namely support vector machine, radial basis function network, decision tree J48, naïve Bayes and Bayesian network. The k-fold cross validation technique is adopted. Computational experiments with the proposed method showed an improvement of 16.14% to 22.49% in the classification accuracy compared to literature survey. Among the ensemble methods tried, MultiBoost ensemble with SVM classifier and grading ensemble with naïve Bayes showed the best performance followed by MultiBoost, stacking and grading ensemble with Bayesian classifier, rotation forest ensemble with RBF and grading and rotation forest ensemble with J48. This investigation conclusively proves the significance of cascading k-means clustering with ensemble methods in the enhanced accuracy in categorization of diabetic dataset. (shrink)

Automatic moving object detection and tracking is very important task in video surveillance applications. In the present work the well known background subtraction model and use of Gaussian Mixture Models have been used to implement a robust automated single object tracking system. In this implementation, background subtraction on subtracting consecutive frame-by-frame basis for moving object detection is done. Once the object has been detected it is tracked by employing an efficient GMM technique. After successful completion of tracking, moving object recognition (...) of those objects using well known Principal Component Analysis, which is used for extracting features and Manhattan based distance metric is used for subsequent classification purpose. The system is capable of handling entry and exit of an object. Such a tracking system is cost effective and can be used as an automated video conferencing system and also has applications like human tracking, vehicles monitoring, and event recognition for video surveillance. The proposed algorithm was tested on standard database on complex environments and the results were satisfactory. (shrink)

This article presents a novel full-reference image quality assessment algorithm by depicting the sub-band characteristics in the wavelet domain. The proposed image quality assessment method is based on energy estimation in the wavelet-transformed image. Image QA is achieved by applying a multilevel wavelet decomposition on both the original and the enhanced image. Next, the wavelet energy and vector are computed to obtain the percentage of the energy that corresponds to the approximation and the details, respectively. Further, the approximate and detailed (...) energy levels of both the original and the enhanced images are compared to formulate an image quality assessment. Numerous experiments are conducted on a dozen of image enhancement algorithms. The results presented show that the image with poor contrast in the foreground than the background has continuous regular coefficient values. The probability density function for such an image has a relatively lower WE and skewness compared with the background. The proposed scheme not only evaluates the global information of an image but also estimates the fine, detailed changes in an enhanced image. Thus, the proposed metric serves as an objective and effective FR criterion for color image QA. The experimental results presented confirm that the proposed WE metric is an efficient and useful metric for evaluating the quality of the color image enhancement. (shrink)

Gearbox is one of the vital components in aircraft engines. If any small damage to gearbox, it can cause the breakdown of aircraft engine. Thus it is significant to study fault diagnosis in gearbox system. In this paper, two deep learning models (Long short term memory (LSTM) and Bi-directional long short term memory (BLSTM)) are proposed to classify the condition of gearbox into good or bad. These models are applied on aircraft gearbox vibration data in both time and frequency domain. (...) A publicly available aircraft gearbox vibration dataset is used to evaluate the performance of proposed models. The results proved that accuracy achieved by LSTM and BLSTM are highly reliable and applicable in health monitoring of aircraft gearbox system in time domain as compared to frequency domain. Also, to show the superiority of proposed models for aircraft gearbox fault diagnosis, performance is compared with classical machine learning models. (shrink)

BACKGROUND AND RATIONALE: Evidence based medicine is the present backbone of rational and objective, modern medical problem solving and is a meeting ground for quantitative and qualitative researchers alike as it culminates into applying the fruits of clinical research to the individual patient. A systematic enquiry into the evolving paradigms in EBM is a need of the hour. AIMS AND METHODS: A qualitative enquiry examining the impact of different methodologies in EBM and their role in generating meaning interpretable at individual (...) levels. RESULTS: Present day outcome based research deals less with patients as individuals than as populations. Evidence based medicine struggles to apply the fruits of population based research to individuals who are often not as predictable as linear quantitative research would like them to be. The present EBM literature neglects a lot of events it doesn't believe to be statistically significant and perhaps here is an area that needs to be improved on - it assumes that because associations are demonstrated between interventions and outcomes in RCTs/meta-analysis, these associations are linear and causal in the real world. While they may be demonstrated repeatedly in highly controlled environments, in the real 'uncontrolled' world of clinical practice with real people, their validity breaks down. CONCLUSIONS: One needs to make the EBM standard model patient-individual (a projection of collective patient event data) resemble the real human individual patient so that optimal EBM individual data that matches our query can be easily and quickly spotted from the dense jungle of information that has grown over the years. This hints at rethinking our entire research methodology and modifying it to suit the needs of the individual patient. (shrink)