International Journal of Computer (IJC - Global Society of Scientific Research and Researchers, GSSRR)
Not a member yet
459 research outputs found
Sort by
Multimedia License Models in a Work of Art When Handling Multimedia Material
Get PDFIn this research, the artist-audience relationship will be studied through the comparison of access control models. These models will be evaluated based on the type of work to be protected, which is artistic creation. A system for creating and interacting with multimedia environments that allow collaboration between artists and the audience is also proposed. This system approaches safety issues in multimedia environments that perform authenticity and watermark mechanisms. The authentication mechanism controls the processes of artists and audiences in multimedia files based on a set of actions required in real-world scenarios. The digital watermarking mechanism handles the protection of copyright and authenticity issues that occur in multimedia systems by applying a secure watermark
Stationary Wavelet Transform(SWT) Based MRI Images Enhancement and Brain Tumor Segmentation
Get PDFBrain tumor is the anomalous growing of Brain cancer cells. Because of its complex structure, brain tumor segmentation and identification are very difficult tasks in medical field. As with MR image processing, MR images are particularly sensitive to noise, resulting in errors in image acquisition and transmission such as Gaussian noise and impulse noise, etc. MRI image is filtered with Median filter and Wiener filter simultaneously to improve the MR image The Stationary Wavelet Transform (SWT) is then used to combine both Median and Wiener filter results. After preprocessing, Adaptive K-means clustering is used for image segmentation. In the post processing step, morphological operation and Median filter are used to get better segmentation results. This method is applied to the BRATS-2015 dataset, which consists of multi-sequence MRI data available to the public from patients with brain tumors. The well-known, based line methods are compared for comparing the proposed system. Mean Square Error (MSE) and Peak Signal Noise Ratio (PSNR) are used in evaluation of the enhancement. For testing tumor segmentation measures, True Positive Rate (TPR), True Negative Rate (TNR), Accuracy, and Jaccard Similarity Index are used. Compared with dependent line methods and state of the art, this system performs well, especially for the entire tumor area
Automatic Paddy Leaf Disease Detection Based on GLCM Using Multiclass Support Vector Machine
Get PDFThe paddy leaf diseases have increased rapidly in the recent years because of globalization, environmental pollution and climate changes which reduce the production of rice and economy of the country. For healthy growth of rice plants there is a need of automatic system which can detect the paddy diseases automatically on time to give the proper treatment for the affected plants. In this paper, we proposed a methodology to develop an automatic system for detect the paddy disease which are Paddy Blast Disease, Brown Spot Disease, Narrow Brown Spot Disease using MATLAB. This paper concentrate on the image processing techniques used to enhance the quality of the image and Multiclass Support Vector Machine to classify the paddy diseases. The methodology involves image acquisition, pre-processing, segmentation, feature extraction and classification of the paddy diseases. Image segmentation technique is used to detect infected parts of leaf by using canny edge detection, multilevel thresholding and region growing techniques. We extract texture features using GLCM (grey level co- occurrence matrix) techniques, additionally we extract color and shape features to improve the accuracy of the framework and use Multiclass Support Vector Machine for classification. We achieved 87.5% accuracy for the test dataset. 
Overlapped Speech Detection in Multi-Party Meetings
Get PDFDetection of simultaneous speech in meeting recordings is a difficult problem due both to the complexity of the meeting itself and the environment surrounding it. The system proposes the use of gammatone-like spectrogram-based linear predictor coefficients on distant microphone channel data for overlap detection functions. The framework utilized the Augmented Multiparty Interaction (AMI) conference corpus to assess model performance. The proposed system offers enhancements over base line feature set models for classification
RWMSI (Read Exclusive Write Exclusive Modified Shared Invalid) Cache Coherence Protocol
Get PDFThis paper proposes a novel coherence protocol RWMSI (Read exclusive Write exclusive Modified Shared Invalid) that merges “snooping and directory – based coherence protocols “and enhanced them depending on the state of “MESI snooping protocol “. “ Coherence” is implemented with “snooping or directory based protocols “. Because of the shared bus the “ Snooping protocols “ are not scalable , while directory protocols incur directory storage overhead , frequent indirections , and are more prone to design bugs
Classification of the Stance in Online Debates Using the Dependency Relations Feature
Get PDFOnline discussion forums offer Internet users a medium for discussions about current political debates. The debate is a system of claims regarding interactivity and representation. Users make claims in an online discussion with superior content to support their position. Factual accuracy and emotional appeal are critical attributes used to convince readers. A key challenge in debate forums is to identify the participants’ stance, each of which is inter-dependent and inter-connected. This research work aims to construct a classifier that takes the linguistic features of the posts as input and outputs predictions for the stance label of each post. Three types of features which include Lexical, Dependency, and Morphology are used to detect the stance of the posts. Lexical features such as cue words are employed as surface features, and deep features include dependency and morphology features. Multinomial Naïve Bayes classifier is used to build a model for classifying stance and the Chi-Square method is used to select the good feature set. The performance of the stance classification system is evaluated in terms of accuracy. The result of stance labels for this proposed research represents as for and against by analyzing the surface and deep features that capture the content of a post
Comparing the Performance of Machine Learning Algorithms for Human Activities Recognition using WISDM Dataset
Get PDFHuman activity recognition is an important area of machine learning research as it has much utilization in different areas such as sports training, security, entertainment, ambient-assisted living, and health monitoring and management. Studying human activity recognition shows that researchers are interested mostly in the daily activities of the human. Mobile phones are used to be more than luxury products, it has become a kind of urgent need for a fast-moving world with rapid development. Nowadays mobile phone is well equipped with advanced processor, more memory, powerful battery and built-in sensors. This provides an opportunity to open up new areas of data mining for activity recognition of human’s daily living. In this paper, we tested experiment using Tree based Classifiers (Decision Tree, J48, JRIP, and Random Forest) and Rule based algorithms Classifiers (Naive Bayes and AD1) to classify six activities of daily life by using Weka tool. According to the tested results Random Forest classifier is more accurate than other classifiers
Using Interactive Charts in a Demonstration Lesson to Help Learners of Colleges of Education Teach Measurement of Angle Properties of Parallel Lines in Basic Schools
Get PDFThe main objective of the study was to examine the efficacy of using interactive charts in a demonstration lesson to equip second year learners of both Foso College of Education and Accra College of Education with practical ways of teaching aspects of mathematics to learners in their placement schools. The topic chosen for the demonstration was measuring angle properties of parallel lines. All second year mathematics major learners in both schools were used for the study. The sample was made up of 4 ladies representing11% and 32 gentlemen representing 89% in Foso College of Education and 12 girls representing 23% and 40 gentlemen representing 77% in Accra College of Education. The data was presented using descriptive statistics. The authors, meeting at a subject based workshop seemed to have the same challenges with their elective mathematics learners in their respective schools and decided to pair and work out strategies to unearth their learners’ creative potentials to teach mathematics in an interactive way. The study identified the causes of the learners’ lack of creative potentials at the pre-intervention stage which was done through observation and interview as inadequate teaching learning resources at the colleges and tutors’ continuous use of lecture method with little or no hands-on activities.
Then the interactive charts designed by Benjamin Tettey were introduced as a suitable option to address the problem. The intervention procedure was carried out by the authors in each other’s college, that is, Miss Mary Acquah in Accra College of Education and Mr. Benjamin Ayerkain Tettey in Foso College of Education. The new pre-tertiary lesson plan template was modified and prepared by Miss Ruby Jecty, a Teacher Professional Learning Coordinator (TPLC) in Foso College of Education, who we engaged in the preparation stage, acted as a critical friend for the observation of the enactments in both colleges and a reflective practitioner who took both tutors through a professional digital reflection to discuss the strengths and areas of improvements of the enactments. A post intervention result established that when learners are exposed to the use of interactive chart intervention activities, they became very imaginative and innovative as they can suggest other innovative ways of teaching other topics in mathematics and develop positive interest towards the teaching mathematics. The results of the study were collated by Miss Mary Acquah. Based on the pre-during and post enactment discussion with the TPLC, the recommendations were made by both tutors. The template is used to format your paper and style the text. All margins, column widths, line spaces, and text fonts are prescribed; please do not alter them
+MRT: Centralized Web Based Application for Managing Repetitive Tasks
Get PDFAbstract—Task management is broadly classified as a process of managing and assigning different tasks to co-workers, employees or subordinates throughout the life cycle of a project via project managers. Tasks can be of various natures but more specifically we are referring here tasks that are used in the software industry. In this modern world, why a project manager has to assign a repetitive task to its team members repeatedly when a new project arrives. This research paper particularly addresses this issue of automatic task management. In this research we have investigated the problems regarding task management, that can occur during the life cycle of a project and presented a web application MRT which can be effectively served as a solution to those problems. Existing applications like Jira and Trello etc. are helping project managers to automate this process. However, we argue that they are still lacking in few aspects. After examining the existing tools, we felt that there are some problems in automatic task management which are not solved yet. We have solved the problems of task management via MRT. We claim that project managers can use MRT for automatic tasks assignments. MRT novel features, therefore, eliminates the need for human effort for repetitive tasks management. Another aspect of Task management is to provide global access to information regarding tasks status to different levels of employees in hierarchy such as manager, CEO, that helps them regarding taking strategic decisions about employees and further projects. MRT can be used as a safeguard against delays and missed deadlines. More precisely our contributions are i) An algorithm for assigning automatic tasks to employees based on their skill set. ii) Tracking of task dependencies that leads to successful project completion iii) Sub Tasks dependencies tracking that are essential to consider, for declaring a task as completed, pending or not completed. iv) Notification management for timely review of tasks. v). Task Tracking at various levels. vi) Priority-based task assignment vii) Additional feature of document storage viii) Creation of Executive Dashboard of the project to visualize the important metrics. ix) Usage of state of the art ASP.Net Web APIs which ensure the use of mobile apps for MRT x) Fresh API Tokens are sent on each call to enhance the security of MRT. xi) The creation of web services that allows integration of MRT data into various devices such as mobiles, tablets etc
The Elderly Fall Detection Algorithm Based on Human Joint Extraction and Object Detection
Get PDFNowadays, the care of the elderly has become a social concern. The fall of the elderly has become one of the main factors threatening the health of the elderly. In this paper, we designed a fall detection algorithm based on human joint extraction and object detection.First,yolov4 was used to identify and detect the elderly. Then openpose was used to detect the human joint. Based on the human joint, this paper using Random Forest to classify the status of the elderly, there are three states of the elderly: falling down, lying down and other states. In the detection of a single old man, the accuracy of the model reached 99.3%, the sensitivity and specificity of the model reached 79.3% and 72.1%