University of Pamplona

Corrosion in steel pipelines can cause critical failures in industrial systems, while conventional inspection methods such as radiography and ultrasonic testing are costly and require specialized personnel. This study presents a mobile computer vision system for automated corrosion detection inside steel pipes using deep learning-based visual analysis. The proposed system consists of a Raspberry Pi 4-based mobile robot equipped with a high-resolution camera for internal inspection. Acquired images were processed using color-space transformations (RGB-HSV), filtering, and segmentation. Convolutional neural networks and semantic segmentation models, including YOLOv8-seg (Instance segmentation) and DeepLabV3 (Semantic segmentation), were trained on a custom corrosion image dataset to identify corroded regions. Real-time visualization was implemented via Flask-based video streaming. Experimental results demonstrated high detection accuracy for uniform corrosion, achieving a mean Intersection over Union (mIoU) above 0.98 and a precision of 0.99 with the YOLOv8-seg model. These results indicate that the proposed system enables reliable and automated corrosion inspection, with the potential to reduce inspection costs and improve operational efficiency. Future work will focus on enhancing real-time performance through hardware optimization.

Vocalization is a fundamental resource for anurans in processes such as territoriality and mating, with intra-and interspecific variation. These signals are modulated by the calls of other species and by the microclimate. In this study, we evaluated vocal activity patterns and their relationship with environmental variables in three artificial ponds in a fragment of tropical dry forest in northeastern Colombia. We propose that biotic and abiotic factors structure vocal activity; we predict that: 1) temporal overlap will tend to be lower between closely related species; 2) dominant frequency will act as a complementary axis of acoustic partitioning, modulating but not determining overlap patterns, and 3) vocal activity will depend on minimum thresholds of temperature and relative humidity, with variation in tolerance among species. We recorded activity during high-and lowprecipitation seasons through continuous acoustic monitoring; we analyzed temporal overlap and applied multivariate clustering with temperature and humidity. We identified 13 species, ten of which showed predominantly nocturnal activity. In Dendropsophus and Leptodactylus, overlap was low, while among species with similar frequencies, it was variable. Vocal activity responded to temperature (>20 degrees C) and humidity (>75%) thresholds, with specific microclimatic ranges for each species. The results show temporal partitioning and a complementary role of dominant frequency, supporting our predictions and highlighting the value of acoustic monitoring for understanding coexistence and community structure in tropical dry forest artificial environments.

This paper addresses the speed control problem of a DC motor in the presence of nonlinearities, disturbances, and unmodeled dynamics by proposing a neural backstepping control scheme based on a Recurrent High-Order Neural Network (RHONN). The proposed RHONN serves as an online approximator to compensate for uncertain nonlinear dynamics in a PD-based backstepping controller, enabling the system to handle disturbances, modeling errors, and unmodeled dynamics. Instead of relying on the traditional Extended Kalman Filter (EKF) for RHONN weight adaptation, the neural parameters are updated online using a Super-Twisting Algorithm (STA). As a result, the proposed STA-based learning law provides a simpler and robust covariance-free adaptation mechanism with practical finite-time convergence properties, making it suitable for real-time embedded implementations. The proposed method was evaluated through numerical simulations and implemented on an embedded microcontroller to assess its real-time performance. Simulation results show reductions between 0.04% and 2.04% in steady-state and integral error metrics compared with a tuned PD controller, and improvements up to 25.66% and 23.82% over LQR and MPC in the IMSE index. Experimental results demonstrate good tracking performance, robustness under varying load conditions, and low computational requirements, confirming the practical feasibility.

This article presents the development of an artificial intelligence-based prototype for training in the surgical procedure of venous cannulation. The system integrates a wearable device equipped with sensors and an anatomical hand dorsum prosthesis designed to simulate five clinical conditions, representing scenarios that healthcare professionals must master with precision and skill. The software incorporates an artificial intelligence model that classifies user performance in real-time into expert or novice categories.This research addresses the need to enhance training in this essential clinical skill, reducing risks for real patients and providing an objective assessment of learning. The findings suggest that this technology serves as a viable alternative to improve medical education and clinical practice within a safe and controlled environment.

Soil acidity severely limits biological activity in high Andean systems, where aluminum toxicity and low pH reduce agricultural productivity and constrain microbial processes. In this context, the effect of calcium carbonate (CaCO3), applied both individually and in combination with a commercial organo-mineral amendment, was evaluated on soil pH, exchangeable aluminum, and microbial respiration in an acidic soil collected from an agricultural farm located in the mountainous region of Pamplona (Norte de Santander), through a 15 day controlled incubation conducted between March and May 2021. A completely randomized design with four treatments (control, 100 % CaCO3, 50 % CaCO3, and 50 % CaCO3 + amendment) and five replicates per treatment was used, measuring chemical variables and respiration at multiple incubation times; statistical analyses included ANOVA, Spearman correlations, and multiple linear regressions. Results showed that CaCO3 significantly increased pH (up to +0.62 units at the full dose) and reduced exchangeable aluminum only in this treatment, while all amended treatments enhanced microbial respiration, particularly during early stages, with a tendency toward a stronger response in the organic combination. However, multivariate analysis revealed that chemical variables did not independently explain respiratory variability, highlighting the predominance of the integrated treatment effect. It is concluded that liming, especially when combined with organic amendments, corrects chemical acidity and revitalizes microbial activity through systemic effects, with practical implications for the sustainable management of high Andean soils.