http://hdl.handle.net/10366/142734 2026-06-11T19:52:05Z http://hdl.handle.net/10366/143317 2019-09-30T00:00:00Z http://hdl.handle.net/10366/143316 An approach to objective assessment of ultrasound examination is presented. To this end, modern information technologies and a set of mathematical methods in the form of a package are proposed. In this paper, diagnosis is viewed as a three-step process, and closed sub-objects are investigated using complex images, which pertains to the earliest diagnostic stage. For this purpose, three new features related to the disclosure of a growth are included in the paper. A system that performs the detection of the growth and finds the coordinates, area, gravity center and color palette of the obtained image is developed. By means of the created software package, the image is cleared from noise, filtering operations are performed, boundaries are defined more clearly and recognition by the mathematical morphology method is completed using selected classifiers. The main purpose is to direct doctor's attention to the presence of the pre-indicator of a non-specific symptom and to control the future development of the growth. The accuracy of the system is confirmed by the detection and identification of closed growths in the images taken in an ultrasound examination of internal organs of the human body. The system's operability has been tested directly on the ultrasound images (138 cases investigated), with the result of 98.8% at the diagnostic stage, 92, 03% at the early diagnostic stage; 2 cases have been recorded at the earliest diagnostic stage in 2018 and the frequency of monitoring has been determined. 2019-12-18T00:00:00Z http://hdl.handle.net/10366/143315 k-Nearest Neighbor (k-NN) is a non-parametric algorithm widely used for the estimation and classification of data points especially when the dataset is distributed in several classes. It is considered to be a lazy machine learning algorithm as most of the computations are done during the testing phase instead of performing this task during the training of data. Hence it is practically inefficient, infeasible and inapplicable while processing huge datasets i.e. Big Data. On the other hand, clustering techniques (unsupervised learning) greatly affect results if you do normalization or standardization techniques, difficult to determine "k" Value. In this paper, some novel techniques are proposed to be used as pre-state mechanism of state-of-the-art k-NN Classification Algorithm. Our proposed mechanism uses unsupervised clustering algorithm on large dataset before applying k-NN algorithm on different clusters that might running on single machine, multiple machines or different nodes of a cluster in distributed environment. Initially dataset, possibly having multi dimensions, is pass through clustering technique (K-Means) at master node or controller to find the number of clusters equal to the number of nodes in distributed systems or number of cores in system, and then each cluster will be assigned to exactly one node or one core and then applies k-NN locally, each core or node in clusters sends their best result and the selector choose best and nearest possible class from all options. We will be using one of the gold standard distributed framework. We believe that our proposed mechanism could be applied on big data. We also believe that the architecture can also be implemented on multi GPUs or FPGA to take flavor of k-NN on large or huge datasets where traditional k-NN is very slow. 2019-08-14T00:00:00Z http://hdl.handle.net/10366/143314 This article aims to discuss the importance of the development of methodologies and control processes by the public power for the proper management of the urban bus system. It reflects on the idea of management control from the perspective of an operational, strategic and innovative perspective. It proposes a new look at the solution of negative externalities that arise in the urban environment of cities such as collisions, roadkill, works that will obstruct the roadbed and various events disrupting local traffic and directly impacting the mobility of public transport on tires. The theme is current, as it addresses the most problematic urban problem in the contemporary world, which is mobility. Thus, the article tells the experience of São Paulo and discusses a new control methodology for urban bus service./n  2019-09-14T00:00:00Z http://hdl.handle.net/10366/143313 Whenever an organization decides to implement an ERP, it must assess its readiness to implement these complex systems. Therefore, the present study aims at considering the pre-implementation phase of ERP and the factors affecting the readiness of the organization for successful implementation of these systems. In this research, the SWOT matrix was used to classify the identified factors. Moreover, fuzzy-dimensional network analysis was used to evaluate decision options due to the weakness of SWOT technique. Since the factors involved in SWOT analysis are not only disjointed, but sometimes there are relationships among some of its factors. Therefore, internal and external factors of the organization are evaluated and prioritized in the research. In addition, finding the strategic position of the organization and identifying the appropriate strategies and prioritizing them to improve the organization's readiness for ERP implementation were other tasks in this research. After analyzing the data, 25 organizational factors are prioritized. The numerical results indicated that the organization is in the offensive position therefore, four strategies are designed and prioritized according to its position. 2019-09-14T00:00:00Z http://hdl.handle.net/10366/143312 The number of flight (NF) is one of the key factors for the administration of the airport to evaluate the apron capacity and airline companies to fix the size of the flight. This paper aims to estimate the monthly NF by performing particle swarm optimization (PSO) and artificial neural network (ANN). Performed PSO-ANN algorithm aims to minimize the proposed evaluation criterion in the training stage. PSO-ANN based on the proposed evaluation criterion offers satisfying fitness values with respect to correlation coefficient and mean absolute percentage error in the training and testing stage. 2019-09-14T00:00:00Z http://hdl.handle.net/10366/143311 Speech Communication in a noisy environment is a difficult and challenging task. Many professionals work in noisy environments like aviation, constructions, or manufacturing, and find it difficult to communicate orally. Such noisy environments need an automated lip-reading system that could be helpful in communicating some instructions and commands. This paper proposes a novel lip-reading solution, which extracts the geometrical shape of lip movement from the video and predicts the words/sentences spoken. An Indian specific language data set is developed which consists of lip movement information captured from 50 persons. This includes students in the age group of 18 to 20 years and faculty in the age group of 25 to 40 years . All have spoken a paragraph of 58 words within 10 sentences in Hindi (Devanagari, spoken in India) language which was recorded under various conditions. The implementation consists of facial parts detection, along with Long short term memory’s. The proposed solution is able to predict the words spoken with 77% and 35% accuracy for data set of 3 and 10 words respectively. The sentences are predicted with 20% accuracy, which is encouraging. 2019-09-14T00:00:00Z http://hdl.handle.net/10366/143310 Face detection is the task of detecting faces on photos, videos as well as the streaming data such as a webcam. Face detection, which is a specific type of general-purpose object detection, is a key prerequisite for many other artificial intelligence tasks such as face verification, face tagging and retrieval, and face tracking. In addition to that, nowadays, face detection is commonly used in daily routines such as social media, and camera software of smartphones. As a result of this necessity, several face detection tools have been proposed. In this study, an experimental performance comparison of well-known face detection tools in terms of (1) accuracy, and (2) elapsed time of detection, which has become even more critical criteria especially when the face detection mechanism is utilized for a real-time system, is proposed. As a result of this experimental study, it is aimed that shed light on the much-concerned query “which face detection tool provides the best performance?”. In addition to that, this study succeeds in showing that convolutional neural networks achieve great accuracy for face detection. 2019-09-14T00:00:00Z http://hdl.handle.net/10366/143309 2019-09-30T00:00:00Z http://hdl.handle.net/10366/143308 Agents deployed to dynamic environments, such as virtual and augmented reality,  need specific mechanisms to capture relevant features from the environment. These mechanisms enable agents to avoid process some useless information and act quickly. The primary goal of this work is to investigate the perception policies of an agent situated in a virtual environment. Perception policies allow giving more priority to sensors perceiving the changes occurring in the environment. Based on the proposed model, each sensor follows a strategy that can change its priority in the overall system. We developed two policies to change the sensors prioritization. The performance evaluation of the proposed model consists of comparing both approaches in a highly dynamic environment. 2019-03-14T00:00:00Z http://hdl.handle.net/10366/143307 India is an agricultural country with an ample amount of arable land that produces wide variety of crops. Growing population and urbanization puts up challenges: more and quality yield in limited area, effective utilization of water resources, inculcating technology with traditional mechanisms, to be faced. A crop irrigation management system with sensor data fetch, transfer and operate functionalities is proposed to meet the expectations. The system comprises of: sensing, data processing and actuator sections, with a network of ambient temperature and humidity at a height and, soil moisture sensor placed at the root zone of the subject. The sensor generated data is compressed and then sent to an FTP server for processing. At the server, a 2-layer Neural Network with 4-Inputs, plant growth, temperature, humidity and soil moisture is used for decision making that controls water supply, fertilizer spray, etc. and a plant is used as the test object. Results show that there is tolerable error in the reconstructed data and 62.5% and 67.5% compression is achieved for ambient temperature, humidity and soil moisture respectively. The decisions are only 2% erroneous when done using Neural Networks using this data. Thus, due to its good data handling, decision making capabilities for precise water usage, being portable and user-friendly, this system proves beneficial in home gardens, greenhouses. 2019-03-14T00:00:00Z http://hdl.handle.net/10366/143306 Online reviews about the purchase of a product or services provided have become the main source of user opinions. To gain profit or fame usually spam reviews are written to promote or demote some target products or services. This practice is known as review spamming. In the last few years, different methods have been suggested to solve the problem of review spamming but there is still a need to introduce new spam review detection method to improve accuracy results. In this work, researchers have studied six different spammer behavioral features and analyzed the proposed spam review detection method using weight method. An experimental evaluation was conducted on a benchmark dataset and achieved 84.5% accuracy. 2019-03-14T00:00:00Z http://hdl.handle.net/10366/143305 Re-engineering (RE) of existing educational institutions (EI) with adoption of latest technology trends (LTT) in form of artificial intelligence (AI) can be great effective in term of quality systems. Increase in student’s strength in class and terrorist attacks on EI urged us to introduce such approach that can assure education quality. Class monitoring with heavy strength always remain major issue for teacher during lecture delivery. In this paper, we implemented reengineering using artificial intelligence based two theories of 1) Multi-face recognition (MFR) system 2) Facial expression recognition (FER) system. Both of these theories supported by intelligent techniques as principal component analysis (PCA), discrete wavelet transform (DWT) and k-nearest neighbor (KNN). After implementation of these intelligent techniques student’s attentiveness will increase. Our developed system can detect expressions like happiness, repulsion, fear, anger, and confusion. Student’s attentiveness score will be displayed on screen. Teacher can interpret on the basis of attentiveness %age. System decision making can be helpful for class continuity or short break. This system is also an application of an expert system (ES) and knowledge base system (KBS) for educational quality assurance. A similar monitoring system was imposed in china with Hikvision Digital Technology. Predations results proved monitoring can be best way for education quality. 2019-05-14T00:00:00Z http://hdl.handle.net/10366/143304 To find out the identical or comparable images from the large rotated databases with higher retrieval accuracy and lesser time is the challenging task in Content based Image Retrieval systems (CBIR). Considering this problem, an intelligent and efficient technique is proposed for texture based images. In this method, firstly a new joint feature vector is created which inherits the properties of Local binary pattern (LBP) which has steadiness regarding changes in illumination and rotation and discrete wavelet transform (DWT) which is multi-resolutional and multi-oriented along with higher directionality. Secondly, after the creation of hybrid feature vector, to increase the accuracy of the system, classifiers are employed on the combination of LBP and DWT. The performance of two machine learning classifiers is proposed here which are Support Vector Machine (SVM) and Extreme learning machine (ELM). Both proposed methods P1 (LBP+DWT+SVM) and P2 (LBP+DWT+ELM) are tested on rotated Brodatz dataset consisting of 1456 texture images and MIT VisTex dataset of 640 images. In both experiments the results of both the proposed methods are much better than simple combination of DWT +LBP and much other state of art methods in terms of precision and accuracy when different number of images is retrieved.  But the results obtained by ELM algorithm shows some more improvement than SVM. Such as when top 25 images are retrieved then in case of Brodatz database the precision is up to 94% and for MIT VisTex database its value is up to 96% with ELM classifier which is very much superior to other existing texture retrieval methods. 2019-05-14T00:00:00Z http://hdl.handle.net/10366/143303 This article presents a hybrid method that uses Convolutional Neural Networks (CNN) to segmentation and Support Vector Machines (SVM) to detection the brandings. The experiments were performed using a cattle branding images. Metrics of Overall Accuracy, Recall, Precision, Kappa Coefficient, and Processing Time were used in order to assess the proposed tool. The results obtained here were satisfactory, reaching a Overall Accuracy of 93% in the first experiment with 39 brandings and 1,950 sample images, and 95% of accuracy in the second experiment, with the same 39 brandings, but with 2,730 sample images. The processing time attained in the experiments was 32s and 42s, respectively. 2019-03-14T00:00:00Z http://hdl.handle.net/10366/143302 This paper addresses the action recognition problem using skeleton data. In this work, a novel method is proposed, which employs five Distance Maps (DM), named as Spatio-Temporal Distance Maps (ST-DMs), to capture the spatio-temporal information from skeleton data for 3D action recognition. Among five DMs, four DMs capture the pose dynamics within a frame in the spatial domain and one DM captures the variations between consecutive frames along the action sequence in the temporal domain. All DMs are encoded into texture images, and Convolutional Neural Network is employed to learn informative features from these texture images for action classification task. Also, a statistical based normalization method is introduced in this proposed method to deal with variable heights of subjects. The efficacy of the proposed method is evaluated on two datasets: UTD MHAD and NTU RGB+D, by achieving recognition accuracies91.63% and 80.36% respectively. 2020-02-14T00:00:00Z