| 1 |
Author(s):
Chidozie Managwu, Ibrahim Kushchu, Daniel Matthias.
Page No : 1-7
|
Browser-Based Object Detection System for Isolating Plastic Bottles using the COCO-SSD Model.
Abstract
Plastic waste, especially in urban environments and water bodies, poses a significant environmental threat. This paper presents a browser-based object detection system for identifying and isolating plastic bottles using state-of-the-art machine learning models. The system leverages TensorFlow.js, ML5.js, and P5.js libraries along with the COCO-SSD model to detect plastic bottles in real time using a mobile camera interface. By employing a browser-based architecture, the system offers cross-platform functionality, eliminating the need for server-based computations or specialized hardware. Experimental evaluation showed high detection accuracy across various environments, underscoring the potential for real-world applications in waste management and recycling efforts.
| 2 |
Author(s):
Umoh Enoima Essien, Sylvester I. Ele.
Page No : 8-26
|
Cuckoo Sandbox and Process Monitor (Procmon) Performance Evaluation in Large-Scale Malware Detection and Analysis.
Abstract
Malware has grown to be an intricate and dynamic threat to cybersecurity. Researchers and cybersecurity specialists use a range of methods to analyze and comprehend malware in order to effectively counter this threat. The malware sandbox is one of the most crucial instruments in this battle. Insights gained by evaluating malware in a sandbox aid in the creation of effective detection. Finding a sandbox that is both highly precise, efficient and affordable is a challenging task. This study compares the effectiveness of Cuckoo Sandbox and Procmon, two of the most popular sandboxes, in the efficient implementation of malware analysis and detection. A Windows 10 Pro window-based computer with a 4 GHz CPU, 16 GB RAM, 8 cores, and a 320 GB hard drive (HDD) was set up. An Oracle virtual machine (VM) for guests was set up and launched. Using the Oracle VM, a virtual operating system (Windows 10 Pro). Furthermore, Yara-Python was deployed and JSON reports, a system built on Python was created. The results show that Cuckoo consistently outperforms Procmon in terms of execution time, completing much more quickly and steadily over each of the ten process runs. Procmon has significantly longer and more fluctuating execution times, peaking at 989 seconds, while Cuckoo maintains execution durations around 530 seconds, suggesting superior efficiency and consistency. Six (6) machine learning-based methods for classifying and detecting malware that used Cuckoo sandbox and process monitor were surveyed. Different performance indicators were found in the six-machine learning-based malware detection and classification studies that Process Monitor was used to survey. A review of six machine learning-based malware detection and classification studies using both Process Monitor and Cuckoo Sandbox indicated that Cuckoo Sandbox consistently delivered better performance. The findings show that machine learning-based malware detection conducted with Cuckoo attained a higher average accuracy of 99.35% compared to 94.48% with Procmon, along with a superior ROC value of 0.97 (97%) versus 0.91 (91%) for Procmon.
| 3 |
Author(s):
Ernest E. Onuiri, Adeyemi John, Kelechi C. Umeaka.
Page No : 27-46
|
MRI-Based Brain Tumour Classification Using Convolutional Neural Networks: A Systematic Review and Meta-Analysis
Abstract
This study systematically reviewed advancements in brain tumor classification using convolutional neural networks (CNNs) with MRI data, aiming to assess the effectiveness and potential for future enhancements. An analysis of 37 studies demonstrated the extensive use of CNN architectures and pre-processing techniques, achieving high classification accuracy rates. However, challenges such as class imbalances and model interpretability were identified. To address these issues, the study recommends further exploration of advanced deep learning techniques, ensemble methods, and the inclusion of more diverse datasets. A maximum accuracy of 98.80% was reported on a dataset comprising 154 MRI brain images, demonstrating the effectiveness of CNNs in brain cancer detection. Additionally, a five-year meta-analysis (2018-2022) on MRI scan cases across different demographic groups revealed important patterns in healthcare resource allocation. This research not only provides a comprehensive evaluation of CNN usage in MRI-based brain tumor classification but also outlines directions for future research to enhance diagnostic capabilities.
| 4 |
Author(s):
Stephen-Orok Duke, Atte Enyinghi Okwong, Emmanuel U. Oyo-Ita.
Page No : 47-57
|
The Synopsis of Soft Computing.
Abstract
The traditional method of problem-solving, known as hard computing, is limited in its ability to handle modern digital technology and real-world problems accurately. Soft computing, a newer paradigm, offers a more versatile approach by utilizing multi-valued logic and human knowledge to solve complex, nonlinear problems efficiently. Unlike hard computing, soft computing can handle imprecise data and uncertainty effectively. This methodology has been successfully applied across various sectors, including scientific, industrial, and medical fields, providing more accurate results. Soft computing is good for its contributions to revolutionizing problem-solving techniques, being tolerant of imprecision, uncertainty, and linguistic variables, and offering approximate solutions to intricate problems.
| 5 |
Author(s):
Ndueso Udoetor, Godwin Ansa, Anietie Ekong, Anthony Edet.
Page No : 58-80
|
Intelligent System for Detection of Copyright-Protected Data for Enhanced Data Security.
Abstract
In the digital era, the proliferation of digital content has intensified concerns over intellectual property rights infringement, highlighting the need for robust copyright protection solutions. This paper presents a software solution designed to address these challenges by combining advanced algorithms with intuitive user interfaces for effective copyright enforcement. Central to the software’s functionality is the Most Significant Bit (MSB) embedding technique, which allows users to imperceptibly embed copyright or trademark information into digital images. This method modifies the MSB of pixel values to encode protection data while maintaining the visual integrity of the images. In the detection phase, the software employs Deep Convolutional Neural Networks (DCNN) to identify instances of unauthorized use or copyright infringement. By analyzing submitted images, the DCNNs use sophisticated pattern recognition algorithms to detect embedded copyright information or trademarks, promptly flagging infringements for further action. The software ensures a seamless user experience with an intuitive interface that guides users through image upload, copyright embedding, and infringement detection processes. This comprehensive approach provides a powerful tool for safeguarding intellectual property rights in the digital landscape, offering users an efficient means to protect and enforce copyright effectively.
| 6 |
Author(s):
Adesoji Adegbola, Akande Oyebola, Tunde-Idowu Inioluwa, Adebanjo Adedoyin, Adewuyi Oluwaseyi, Mgbeahuruike Emmanuel, Adediran Oluwaseyi.
Page No : 81-93
|
Development of a Web-Based Document Repository with Plagiarism Checker.
Abstract
In the digital age, managing vast volumes of documents and ensuring the originality of content has become a significant challenge. This paper presents the development of a web-based document repository integrated with a plagiarism checker, aimed at providing a comprehensive solution for storing, retrieving, and verifying the uniqueness of documents. The system allows users to upload, organize, and search documents efficiently while employing a robust plagiarism detection mechanism to ensure the integrity of content. By leveraging web technologies and plagiarism detection algorithms, this platform serves as a valuable tool for educational institutions, businesses, and content creators. The system enhances document management practices by offering a centralized, secure repository and reducing the risk of intellectual property infringement. This paper discusses the architecture, features, and implementation challenges of the system, along with its potential applications in various domain
| 7 |
Author(s):
Adesoji Adegbola, Akande Oyebola, Mgbeahuruike Emmanuel, Adebanjo Adedoyin, Adewuyi Oluwaseyi.
Page No : 94-102
|
A Convolutional Neural Network Model for Crop Disease Detection System.
Abstract
Crop diseases pose a significant challenge to global food security, adversely impacting agricultural output and resulting in considerable economic repercussions. The prompt and precise identification of these diseases is essential for effective intervention and sustainable agricultural practices. This study introduces a model based on Convolutional Neural Networks (CNNs) for the automated detection of crop diseases. The model employs advanced deep learning methodologies to recognize and categorize plant diseases through the analysis of leaf images. Our CNN framework is trained on an extensive dataset comprising both diseased and healthy plant images, employing multiple convolutional layers to extract intricate features, including texture, color variations, and patterns linked to specific diseases. The model demonstrates a high level of accuracy in identifying a variety of diseases across different crop species by learning from both overt symptoms and subtle cues. We evaluate the performance of the system using established metrics such as accuracy, and precision, thereby validating its efficacy in practical applications. The proposed system is designed for implementation in low-resource agricultural settings, offering farmers a cost-effective, dependable, and real-time solution for monitoring crop health.
