International Journal of Innovations in Science & Technology
Not a member yet
813 research outputs found
Sort by
Detection of Holes in Point Clouds Using Statistical Technique
Get PDFA point cloud is a dynamic, three-dimensional geometric representation of data that has different qualities for every point, including geometry, normal vectors, and color. However, holes that often occur during the 3D point cloud collection process provide an immense obstruction to the analysis and reconstruction of point clouds. Thus, detecting these holes is a crucial initial step toward obtaining precise and comprehensive representations of the real surfaces. Although there are several methods available for hole detection and filling, the problem is exacerbated by their shortcomings, which include high computation complexity or limited effectiveness. Our method is based on a sequence of basic but efficient statistical techniques. Our method is based on a sequence of basic but efficient statistical techniques. First, we find the mean distances between each point using the K Nearest Neighbors (KNN) technique. Next, we can categorize normal points and points that belong to holes and borders by using this mean as a threshold. Our method\u27s simplicity and low computational resource needs offer significant advantages over other approaches
An Aggregated Approach Towards NILM on ACS-F2 Using Machine Learning
Get PDFThe Energy Sector across the globe is experiencing rapid growth, driven by Internet of Things (IoT) integration technologies and advanced algorithms. This evolution is particularly evident in the ongoing competition among tech companies in the development of smart metering solutions. Despite these advancements, a critical challenge persists— the lack of definitive technical protocols for monitoring the total usage or power signatures of individual appliances, referred to as non-intrusive load monitoring (NILM) in aggregate. While intrusive load monitoring (ILM) provides very accurate and thorough insights, non-intrusive methods are essential to address losses specially in residential areas. In this research a groundbreaking approach is proposed towards handling NILM problems by analyzing and aggregating the load patterns of four key appliances of daily use, namely the Coffee Machine, Fridge, Kettle, and Laptop from the ACS-F2 dataset. The generated aggregated dataset, is systematically combined using electrical formulations to yield the desired data which reflects the simultaneous operation of multiple appliances, this has been explored for the first time in the known literature. The proposed dataset contains around 6750 aggregated appliance load patterns for both training and testing. Furthermore, multiple Time Series Classifiers (TSC) were gauged using a suite of evaluation metrics, on the proposed dataset and an accuracy of 92.1% was achieved by the CATCH22 classifier
Stress Detection and Prediction Using CNNs from Electrocardiogram Signals
Get PDFStress prediction is a crucial aspect of mental health monitoring, with consequences for both psychological well- being and productivity. This work presents a unique way for stress prediction that uses binary and multiclass classification models. Through extensive experimentations with different durations and frequencies of Electrocardiogram Signal (ECG) signals, we identified a 5-second dataset sampled at 200Hz as the optimal configuration for our model. Moreover, we introduced an innovative feature i.e., the prediction of stress scores ranging from 0 to 100, providing nuanced insights into stress levels, where 0 represents no stress and 100 indicates high stress levels. The model obtains 95.04% accuracy, 95.27% precision, 94.95% F1 score, 86.69% sensitivity, and 99.44% specificity for the binary classification. With "Fun" added to the list of stress categories in addition to "Base" and "TSST," the model continues to perform well in the multiclass classification scenario, with accuracy of 88.10%, precision of 87.60%, F1 score of 87.35%, sensitivity of 95.97%, and specificity of 79.23%. These findings highlight how well this applied strategy predicts stress levels, providing important information for mental health and stress management strategies
Automated Seismic Horizon Tracking Using Advance Spectral Decomposition Method
Get PDFIntroduction/Importance of Study: In three-dimensional seismic interpretation, automatic horizon tracking is a critical productivity tool. However, it often fails in areas where horizons are not smooth and exhibit sharp discontinuities such as large spatial displacement or changes in reflector aliasing, horizon gradients, and signal character. Such failures require manual intervention, which increases the interpretation cycle time.
Novelty Statement: In this research study, an automated horizon tracker is proposed that adapts to changes in reflector shape, strength, and geological variation as it traverses through the seismic data volume.
Material and Method: A predefined spatial grid window steers across the horizon surface where its orientation changes with the variation in a pre-computed, high-resolution, dip volume. The method is further improved to incorporate tracking horizons across discontinuities i.e. faults.
Result and Discussion: The proposed method is tested on three-dimensional seismic data with varying geological conditions and has demonstrated successful mapping of horizon surfaces and effective matching across major faults.
Concluding Remarks: Our automatic procedure, by reducing the need for manual intervention during interpretation, has the potential to significantly improve productivity
Evaluating the Effectiveness of Phase Difference in Early Drought Detection
Get PDFIntroduction. This research work focuses on how various phase relationships can enhance our understanding of the effects of drought on moisture deficiency in desert ecosystems, an extensive and damaging environmental phenomenon that affects natural ecosystems, economies, health, agriculture, and society.
Novelty Statement. The primary objective of this research is to inspect the lag time variance between fixed and dynamic lag windows correlated with NDVI, aiming to devise an optimal methodology for drought analysis in this region.
Material and Methods. Leveraging remote sensing data, this study delves into the complex drought dynamics of the Thar Desert, employing a comprehensive analysis of 22 years of CHIRPS rainfall time series data and MODIS NDVI (Normalized Difference Vegetation Index) product. This study performed a cross-correlation of rainfall and NDVI, comparing the lag time difference between fixed lag windows (16, 32, 48, 64 days) and dynamic lag windows (ranging from 4 to 64 days with incremental steps) against 22 years of NDVI data of MODIS.
Results and Discussions. The preliminary results showed that dynamic lag windows of 4, 8, 12, 16, …, and 64 days exhibit the highest correlation with NDVI, with a lag time of 40 days showing maximum correlation. These findings suggest that dynamic lag windows capture the temporal variability of drought impact on vegetation more effectively compared to fixed lag windows in the Thar Desert. The same work was done with a sub-dynamic lag window ranging in between the highly correlated lag episodes of dynamic and fix windows respectively i.e.,40 days and 48 days, concluding that a lag phase of 42 days exhibits the highest correlation with vegetation more effectively. Furthermore, the study unveils a significant drought event in 2002, showcasing the sensitivity of the dynamic lag approach in detecting extreme drought occurrences.
Concluding Remarks. This research not only advances drought analysis methodologies in arid regions but also underscores the imperative for future investigations to explore the generalizability of dynamic lag windows across diverse regions and evaluate their predictive capacity in forecasting drought-induced vegetation changes
Assessment of Three Fast Growing Populus Deltoides Species in Various Soil Profiles Under Nursery Conditions
Get PDFPopulus plants are fast-growing plants exhibiting strong adaptability and a short rotation period, with enhancing ability of carbon stock, helping in combating climate change and sustaining livelihoods. Pakistan has a shortage of firewood and timber. Thus, hybrid fast-growing plants are the only way to balance wood demand and supply in country. Therefore, objective of the present study was to evaluate and compare the growth patterns and carbon stocks of Populus deltoides varieties in soil media under nursery conditions. To achieve this, three fast-growing hybrid species of Populus deltoides, Italian Populus (euramerciana), clone A-Y48, and local Populus were used. Three healthy plants of mothers aged one to two years were selected from field area of the Rangeland Research Institute, NARC. The cuttings were planted in 90 pots after being filled with three different media, and plant growth was recorded after seven days for the number of leaves, height, diameter, and irrigations frequency applied to each pot. Three-month data were collected and analyzed by using an RCBD design. Afterwards, all the plants were harvested, and soil samples were taken from the pots and brought to the RRI laboratory for estimation of total biomass and carbon stocks. It was concluded that Clone AY-48 achieved highest height among all Populus deltoides varieties and stored more carbon stock in comparison to Italian and local poplar varieties. Farmyard manure had a positive influence on height of the different Populus deltoides varieties. Clone AY-48 and Italian poplar plants are more suitable for rapid growth
Designing Flood Risk Reduction Plan for Kalat Division, Balochistan
Get PDFFlood risk mitigation is crucial in the Kalat Division of Balochistan Province due to frequent flooding events that endanger lives and infrastructure. This study introduces a novel approach by integrating Landsat 8-9 OLI data with advanced remote sensing techniques to address flood risks in the region, an approach not previously utilized. Covering the period from 2015 to 2022, the research employs satellite imagery and indices such as NDWI, MNDWI, NDVI, LULC, and Watershed Analysis, with thorough pre-processing to ensure data accuracy. NDWI and MNDWI analyses effectively mapped and monitored water bodies, pinpointing vulnerable areas essential for flood risk assessment. NDVI analysis revealed significant correlations between vegetation dynamics and flooding, highlighting ecological impacts. LULC analysis identified substantial changes in land use patterns, emphasizing the role of human activities in flood vulnerability. Watershed Analysis offered valuable insights into hydrological dynamics and precipitation patterns, supporting flood prediction and mitigation efforts. This integrated approach provided a comprehensive understanding of climatic, hydrological, and land cover factors contributing to flood vulnerability, enabling the development of evidence-based flood risk management strategies. The findings enhance the Kalat Division\u27s resilience against future floods through informed, evidence-based mitigation strategies
A Review based on Active Research Areas in Mining Software Bug Repositories: Limitations and Possible Future Trends
Get PDFIntroduction/ Importance of Study: Bug repository mining is a crucial research area in software engineering, analyzing software change trends, defect prediction, and evolution. It involves developing methods and tools for mining repositories, providing essential data for bug management.
Objective: The goal of this study is to analyze and synthesize recent trends in mining software bug repositories, providing valuable insights for future research and practical bug management.
Novelty statement: Our research contributes novel insights into mining software repository techniques and approaches employed in specific tasks such as bug localization, triaging, and prediction, along with their limitations and possible future trends.
Material and Method: This study presents a comprehensive survey that categorizes and synthesizes the current research within this field. This categorization is derived from an in-depth review of studies conducted over the past fifteen years, from 2010 to 2024. The survey is organized around three key dimensions: the test systems employed in bug repositories, the methodologies commonly used in this area of research, and the prevailing trends shaping the field.
Results and Discussion: Our results highlight the significance of artificial intelligence and machine learning integration in bug repository mining; that has revolutionized software development process by enhancing classification, prediction and vulnerability detection of bugs.
Concluding Remarks: This survey aims to provide a clear and detailed understanding of the evolution of bug repository mining, offering valuable insights for ongoing advancement of software engineering
Design of a High Gain Dual Band Patch Antenna with T Slot Ground Structure for Millimeter Wave Communication Applications
Get PDFThis paper introduces a novel design approach for achieving high gain, dual-band operation, and enhanced bandwidth in a microstrip patch antenna tailored for 5G applications. The antenna operates at the millimeter-wave bands of 28 GHz and 38 GHz, crucial frequencies for the next-generation 5G wireless communication systems. The proposed design employs two inverted T-shaped slots on the patch to enable dual-band functionality. Simultaneously, a very high gain is attained by strategically inserting two inverted T-shaped slots on the radiating element of the patch. To further improve the antenna\u27s bandwidth, a ground slot structure with three different types of slots U-shaped, L-shaped, and T-shaped are compared on the ground plane. The best bandwidth enhancement is achieved by T shape Slot on both bands. The substrate chosen for the antenna fabrication is Rogers RT Duroid 5880, characterized by a thickness of 0.501mm, a low loss tangent of 0.0009, and a relative permittivity constant of 2.2. The simulations are conducted using Ansys HFSS software proposed antenna design, demonstrate impressive performance metrics. Maximum gains of 17 dB at 28 GHz and 38 GHz are achieved form T shape slot ground configuration, the U-shaped slot configuration yields a maximum gain of 15 dB, and the L-shaped slot configuration achieves 7.8 db. Furthermore, the impedance bandwidth response at the respective resonating frequencies extends to 1 and 2 GHz below the -10dB line, showcasing the antenna\u27s excellent bandwidth characteristics. In terms of form factor, the proposed antenna is compact, measuring 16.2 x 12.8 x 0.501 mm. This compact size, coupled with the high gain and wide bandwidth at both operating bands, making the antenna well-suited for integration into 5G applications
PERFORMANCE ANALYSIS OF A HYBRID RECOMMENDER SYSTEM
Get PDFIn the prevailing information age, human confrontation with extensive information makes it difficult to segregate the relevant content on the basis of choices and priorities. This gives rise to the need for effective recommendation systems that can be incorporated into distinct and diversified domains such as e-commerce, social media, and news media websites and applications. By giving suggestions, these recommender systems efficiently reduce huge information spaces and direct the users toward the items that best match their requirements and preferences. Hence, they play an important role in filtering out the relevant user-specific information. Based on the working principle, recommender systems can be classified into Content-Based Systems, Collaborative Filtering Systems, or P opularity-Based Systems. However, to cope with the problems of cold-start and plasticity that are associated with standalone recommender systems, hybrid recommendation systems are being introduced. This research is therefore focused on the development of a Weighted Hybrid Model that combines the scores of the three standalone recommender models in a linear fashion. The performance of the proposed hybrid model is tested against all three standalone models on an online News dataset. Using a Top-N accuracy metric, it is found that the accuracy of the weighted hybrid model is higher than the standalone Content-Based, Collaborative, and Popularity-Based models against the same dataset. An efficiency of 90% for the Hybrid model was achieved compared to the best-performing standalone model having an efficiency of 53%