Our Work

Abstract: A machine learning-based web application for maize leaf disease classification was developed to facilitate the early detection and classification of diseases affecting maize plants. The syst…

Abstract: A machine learning-based web application for maize leaf disease classification was developed to facilitate the early detection and classification of diseases affecting maize plants. The system employs the Random Forest algorithm to classify maize leaf images into four categories: Healthy, Gray Leaf Spot, Common Rust, and Blight. The training dataset consists of 1,006 images of Common Rust, 574 images of Gray Leaf Spot, 1,146 images of Blight, and 1,162 images of Healthy leaves, ensuring effective model training and classification. The trained Random Forest model was saved and integrated into a Python Flask-based web application, enabling users to upload images of maize leaves for real-time disease prediction. Upon image upload, the system accurately identifies the disease or confirms the leaf's healthy condition, providing quick and reliable results. This web-based tool offers farmers, agricultural professionals, and researchers an accessible solution for the timely detection and management of maize leaf diseases, thereby supporting improved crop management and disease control. The proposed model achieved an accuracy of 83% during testing, demonstrating its effectiveness for maize leaf disease classification.

This record is Data product 4 of GHM_drought and provides individual-model EDDI projections for 2025–2100 from 16 bias-corrected CMIP6 models under SSP1-2.6, SSP2-4.5, and SSP5-8.5. Data product…

This record is Data product 4 of GHM_drought and provides individual-model EDDI projections for 2025–2100 from 16 bias-corrected CMIP6 models under SSP1-2.6, SSP2-4.5, and SSP5-8.5. Data product 1 provides observation-based historical drought indices for 1961–2024, together with multi-model ensemble-mean projections and inter-model uncertainty ranges for 2025–2100, while Data products 2 and 3 provide the corresponding individual-model SPEI and SPI projections, respectively. Associated records: Data product 1, https://doi.org/10.5281/zenodo.18045718; Data product 2, https://doi.org/10.5281/zenodo.21491403; Data product 3, https://doi.org/10.5281/zenodo.21523380; Data product 4, https://doi.org/10.5281/zenodo.21540518.   1. Description Dataset name: GHM_drought Summary: The GHM_drought dataset is a new Global Meteorological Drought Dataset at 0.5° spatial resolution for the period 1961–2100, derived from the Climatic Research Unit (CRU) dataset and 16 bias-corrected Coupled Model Intercomparison Project Phase 6 (CMIP6) models. This dataset calculates Standardized Precipitation Index (SPI), Evaporative Demand Drought Index (EDDI), and Standardized Precipitation Evapotranspiration Index (SPEI) based on a unified framework, ensuring consistency and comparability among the indices. Additionally, the dataset provides multiple accumulation timescales (1, 3, 6, 9, 12 months, and 1 year) and multiple scenarios, including historical (1961–2024) and future Shared Socioeconomic Pathway (SSP) scenarios (2025–2100) (SSP1-2.6, SSP2-4.5, and SSP5-8.5). Crucially, the dataset provides uncertainty ranges for future projections to enhance reliability. Validation results demonstrate that the GHM_drought captures historical drought events robustly and maintains high consistency with existing benchmark datasets. For future projections, the applied threshold-based quantile mapping method effectively corrects systematic biases. The GHM_drought aids global drought monitoring and projection, thereby supporting climate risk assessment and adaptation. Latest version: Version 1 (Jul. 25, 2026)   2. Content of the dataset This dataset contains single-model Evaporative Demand Drought Index (EDDI) data at six accumulation timescales: 1 month, 3 months, 6 months, 9 months, 12 months, and 1 year. EDDI_{accumulation timescale}.zip: Each accumulation-timescale archive contains 48 EDDI datasets in NetCDF format, comprising 16 CMIP6 models under three future scenarios: SSP1-2.6, SSP2-4.5, and SSP5-8.5. Each NetCDF file represents the EDDI output from one individual CMIP6 model under one future scenario for the period 2025–2100. The files are named according to the following convention: EDDI_{accumulation timescale}_{scenario}_{model}.nc For example: EDDI_3-month-scale_SSP245_ACCESS-CM2.nc and EDDI_1-year-scale_SSP585_MPI-ESM1-2-HR.nc   3. Details of the variables in the files Each single-model EDDI NetCDF file contains the following four variables: (1) lat: Latitude coordinate, measured in degrees (°). (2) lon: Longitude coordinate, measured in degrees (°). (3) time: Time coordinate. (4) eddi: Evaporative Demand Drought Index variable with dimensions (time, lat, lon).   4. Examples of utilization The NetCDF files can be accessed and processed using various software tools, including GIS applications such as ArcGIS Pro, visualization tools like Panoply, and programming libraries such as xarray in Python.

Submission artifact for the manuscript SD-I/O: Actuation-Gated Runtime Reconfiguration for Embedded Control Nodes. The artifact contains retained E1--E7 experiment logs, available failed or partial ru…

Submission artifact for the manuscript SD-I/O: Actuation-Gated Runtime Reconfiguration for Embedded Control Nodes. The artifact contains retained E1--E7 experiment logs, available failed or partial runs, E3 controller sources and WebAssembly artifacts, host-side scripts, firmware source snapshot, recovered build/configuration evidence marked as not scenario-bound, metadata, checksums, and known evidence-boundary notes.

El yo ESTROMÁTICO también tiene en cuenta el entorno de cada época histórica .

Abstract: The productivity of tomato crops is negatively impacted by a number of leaf diseases, including Yellow Leaf Curl Virus, Septoria Leaf Spot, Early Blight, and Late Blight. In rural areas, man…

Abstract: The productivity of tomato crops is negatively impacted by a number of leaf diseases, including Yellow Leaf Curl Virus, Septoria Leaf Spot, Early Blight, and Late Blight. In rural areas, manual disease detection is frequently unreliable and unavailable. This study presents LeafLens, a lightweight hybrid tomato leaf disease detection framework integrating CNN-based learning with statistical texture descriptors for practical agricultural deployment with 93.5%. The model uses deep learn-ing and lightweight classifiers to classify leaf images after extracting spatial, colour, and texture features. The scalable, affordable, and user-friendly design of LeafLens helps to increase agricultural sustainability and productivity.

Videos supporting the deliverable report D4.1 - EndoFLight bioprinting unit design and development of the LUMINATE project. The videos include: D4.1 Video 1.mp4 and Extrusion from syringe + Endoscopi…

Videos supporting the deliverable report D4.1 - EndoFLight bioprinting unit design and development of the LUMINATE project. The videos include: D4.1 Video 1.mp4 and Extrusion from syringe + Endoscopic crosslinking 2.mp4: preliminary experiments showing the FLight crosslinking process in a bovine knee ex vivo model

This dataset contains a documented Microsoft Excel workbook for a conceptual SeaLink Highway engineering and cost model. The workbook includes wind and wave loading calculations, structural and founda…

This dataset contains a documented Microsoft Excel workbook for a conceptual SeaLink Highway engineering and cost model. The workbook includes wind and wave loading calculations, structural and foundation feasibility checks, 1,700 km and 40 km CAPEX estimates, documented input assumptions, documented cost parameters, a data dictionary, quality-control notes, and a one-at-a-time sensitivity analysis. No MATLAB or Python code was used. All calculations and the sensitivity analysis are contained in the Microsoft Excel workbook. The model is intended for conceptual feasibility assessment only and does not represent a construction-ready design, tender estimate, or site-specific engineering assessment.

The rise of various social media platforms and digital media contents, such as political memes, significantly impact the democratic processes within a country. In seeking to determine their capability…

The rise of various social media platforms and digital media contents, such as political memes, significantly impact the democratic processes within a country. In seeking to determine their capability in raising awareness, the researchers conducted a study aimed to determine the extent exposure to political meme culture as a political discourse; the level of socio-political awareness in terms of political information, societal events, and political interest; and the relationship between the extent exposure to political meme culture as a political discourse and socio-political awareness among selected youth. The study was conducted from March 2025 to May 2025 in selected barangays in the City of Imus, Cavite which are Buhay na Tubig, Carsadang Bago II, and Pasong Buaya II. Specifically, this study utilized Quantitative descriptive-correlational approach to effectively identify the relationship of the variables. The authors used a self-made survey questionnaire to gather data. Three hundred eighty-four (384) were purposively chosen in line with the criteria of the study. The results revealed that the respondents have frequently encountered political meme culture which implies that they are moderately exposed to political meme culture but not always. It was also found that the respondents have a high level of socio- political awareness in terms of political information, societal events, and political interest. However, Pearson R revealed that these two variables have no significant relationship.