Ingeniería en Sistemas, Electrónica e Industrial

URI permanente para esta comunidadhttp://repositorio.uta.edu.ec/handle/123456789/1

Examinar

Resultados de la búsqueda

Mostrando 1 - 10 de 35

Clasificación de residuos mediante una aplicación android con un modelo entrenado para el aprendizaje en niños de primaria
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software., 2026-07-07)
Peña López, John Alejandro
;
Ibarra Torres, Oscar Fernando
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas Electrónica e Industrial. Carrera de Software.
This research addresses the problem of incorrect waste classification through the development of "MiniEco", an Android mobile application integrated with artificial intelligence for primary school students at the Juan León Mera La Salle Educational Unit. The Mobile-D methodology was used for software development and CRISP-DM for artificial intelligence. After initially training a ResNet50 architecture with TrashNet, it failed in real-time due to unvaried backgrounds and camera frame processing issues. Therefore, the system was optimized using a more diverse, real world dataset and selecting the EfficientNetV2S model. This model was converted to TensorFlow Lite to ensure 100 percent offline operation, achieving a laboratory accuracy of 94.50 percent with a final application size of 36 MB, which maintained a CPU consumption of 16 percent on low-end hardware. In the validation phase with 32 students, a significant increase in ecological knowledge was recorded, moving from 18.75 percent correct answers in the initial diagnosis to an effectiveness of up to 93 percent in tests with the tool. The usability evaluation using the SUS metric showed a satisfaction rate of 96 percent among the students. It is concluded that the integration of local computer vision and mobile gamification strengthens environmental learning at an early age, overcoming institutional connectivity limitations and providing an autonomous, scalable, and highly efficient pedagogical tool.
Sistema de detección de niveles de atención en estudiantes mediante análisis de imágenes de video empleando visión artificial
(2025-08)
Yugcha Rubio Edison David
;
Castro Martin Ana Pamela
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones
The present work developed a system for real-time monitoring of the attention level of students, through the implementation of computer vision and deep learning. The system uses a webcam to capture images of the student, which are processed by algorithms such as MediaPipe and YOLOv8 to detect postures, gestures, gaze direction and distracting objects. From this data, a neural network-based model classifies the level of attention into three categories: high, medium and low. The solution was validated at the Gustavo Iturralde Fiscal School, located in the Lasso parish, where structured tests were applied to 12-year-old students. The results showed adequate accuracy and significant agreement between the system's automatic evaluation and the teacher's direct observation. In the activities carried out, the system and the observer coincided in the classification of attention levels, consistently identifying high, medium and low states according to the behavior of the students during the different tests. This alignment between the two methods reinforces the usefulness of the system as a complementary tool for the objective monitoring of attention in the classroom. The implementation proved to be practical, non-invasive and compatible with the school environment. The system provides a valuable tool for improving pedagogical management, allowing teachers to identify attention levels in a timely manner and apply corrective strategies. Its adaptability and accuracy position it as a technological resource with high potential for future research and educational applications.
Sistema de clasificación automático de fresas según su calidad empleando inteligencia artificial
(2025-07)
Manobanda Jimenez Kevin Andres
;
Castro Martin Ana Pamela
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones
Strawberry cultivation in Ecuador has significant economic and social relevance. Thanks to their quality, nutritional value, and culinary versatility, the country has positioned itself as an exporter to the United States, Europe, and Latin America, markets that demand high quality standards and advanced technologies. This project proposes an automatic strawberry classification system using artificial intelligence, structured in four stages. The first stage consisted of acquiring images with the rear camera of a Samsung A10 smartphone. These images were labeled using Roboflow and organized into a database. In the second stage, a YOLOv11l model was trained in Google Colab, and Python and libraries such as tkinter (graphical interface), PIL (image visualization), OpenCV (video processing), ultralytics. YOLO (AI model), serial (communication with Arduino), and time to run the AI model, classifying the strawberries into “Extra,” “First,” “Second,” and “Discard.” For efficient system execution, an HP OMEN 15 laptop was used, taking advantage of its NVIDIA GeForce GTX 1660 Ti graphics card, which significantly improves the model's processing speed. The third stage included the construction of an automated sorting machine with Arduino Mega, actuators, and buttons for the physical separation of strawberries. Finally, a user-friendly interface was developed to display results in real time. The system achieved 83.05% sorting efficiency, reducing human error, optimizing processes, and ensuring uniform quality control. This improves the competitiveness of Ecuadorian producers in local and international markets.
Sistema de monitoreo y detección de comportamiento delincuencial aplicado en medios de transporte utilizando visión artificial
(2025-07)
Abril San Lucas Jonathan Ronaldo
;
Guamán Molina Jesús Israel
;
Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones
Nowadays, the different video surveillance systems within the means of transport are largely dependent on monitoring and human control, and as a result, the costs incurred by the owners of the means of transport are high to carry out this objective. In a way, a proposal has been made for a monitoring and detection system of delinquent behavior applied to means of transport using artificial vision. The system is implemented inside an NVIDIA Jetson Orin Nano with a 720p webcam and a Neo 6m GPS module for georeferencing. The system uses libraries such as OpenCV and MediaPipe for video processing and body posture estimation, integrating AI models (YOLO and SVM) for the detection of suspicious objects and threatening postures. Events are classified as “Normal”, ‘Suspicious’ or “Flagrant Crime” based on combined thresholds. Alerts, containing captures and coordinates, are sent in real time to Slack and managed from a Flask web interface. Absolutely all detections are stored within a SQLite database for traceability and future enhancements. This system automates security monitoring, improving responsiveness and crime prevention.
Sistema de gestión inteligente del parqueadero del edificio de Bienestar Estudiantil en la Universidad Técnica de Ambato mediante visión artificial
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2025-02)
Gavilanez Jiménez Marlon Abel
;
Guamán Molina Jesús Israel
This thesis project presents a smart management system for the parking lot of the student welfare building at the Universidad Técnica de Ambato using computer vision. To capture real-time images of the parking lot, two IP cameras are installed at the entrance and exit, with processing performed on the NVIDIA Jetson Nano development platform. The system employs the YOLOv8 computer vision algorithm for vehicle detection and counting. It monitors vehicle flow by recording entries and exits to maintain an updated count of the total occupancy of the parking lot, which has 166 spaces. The analysis results are displayed through a graphical interface developed in Python using Tkinter, running on an Ubuntu environment. This interface allows staff to monitor the number of vehicles in the parking lot and the remaining availability in real time.Finally, the collected data is stored in a local database, enabling historical tracking of vehicle flow and facilitating analysis for optimizing parking lot management.
Identificación temprana de presencia de plagas en cultivos de ambiente controlado empleando visión artificial y deep learning
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2025-02)
Lascano Villafuerte Erick Fernando
;
Córdova Córdova Edgar Patricio
The cultivation of crops in controlled environments offers optimal conditions for their growth. However, the presence of pests can adversely impact both the quantity and the quality of the produce. This phenomenon exerts a deleterious effect on the economic viability of farmers. To address these challenges, farmers have adopted technological solutions, such as Agriculture 5.0, to enhance their productivity and quality of produce. A study was conducted with the objective of implementing a system for pest detection, utilizing Computer Vision and Deep Learning technologies. It is imperative to detect pests in crops at an early stage to avert production losses. Consequently, the system is predicated on a neural network capable of accurately detecting various pests. The system is comprised of four distinct stages: acquisition, training, processing, and visualization. In the initial acquisition stage, four cameras were utilized to capture images and video. The training stage entailed the utilization of collected data in conjunction with a model adept at functioning with constrained resources while maintaining optimal detection accuracy. The image processing stage entailed the utilization of a microcomputer that had been optimized to operate in conjunction with artificial intelligence. The visualization and information management stage involved the development of a graphic interface capable of displaying the data obtained. The trained model demonstrated an accuracy of 95.7% in the detection of pests, and subsequent system tests yielded a reliability of 93.7%, thus confirming the system's reliability.
Sistema para detección de automóviles robados empleando visión artificial en vehículo aéreo no tripulado (UAV)
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2025-02)
Chanahuano Azogue Jose Luis
;
Santo Taipicaña Danny Gabriel
;
Salazar Logroño Franklin Wilfrido
The present research focuses on the design and implementation of an advanced technological system for the detection of stolen cars, integrating various technologies such as computer vision, YOLOv8 convolutional neural networks, optical character recognition (OCR) and unmanned aerial vehicles (UAV). This innovative system captures, processes and analyzes images of vehicles in parking lots, allowing license plates to be identified and compared with cloud databases managed in Azure. The results obtained are notified through a web platform and applications such as Telegram and email, thus facilitating an agile and efficient response. The system is composed of a UAV equipped with an FPV camera that transmits real-time video to a receiver connected to a computer. The processed information is stored in a database and automatically verified. The UAV configuration allows stable and safe flights, while the artificial intelligence models used guarantee high accuracy in license plate detection, even under variable lighting conditions. In addition, field tests were carried out in different locations to evaluate the system's performance in terms of flight time, energy consumption and data processing. Among the most outstanding results is an 82.78% reliability rate in the detection of vehicle license plates with average information sending times of 20.26 seconds per detected event and sending of respective notifications. This project represents a significant contribution to strengthening public security by providing scalable and technologically advanced.
Sistema de detección temprana de eventos delictivos en entidades comerciales con visión artificial y deep learning
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2025-02)
Oyaque Moncayo Christian Manuel
;
Córdova Córdova Edgar Patricio
Presently, video surveillance systems in commercial entities depend considerably on human monitoring, which results in increased costs for these entities. Consequently, a system for early detection of criminal events using computer vision and deep learning has been proposed. This system employs two complementary processing methods: the first utilizes YOLOv8 for suspicious object detection, and the second employs an SVM algorithm that classifies key points extracted with MediaPipe for threat posture detection. The system operates on an Nvidia Jetson Nano module, which processes videos in real time and displays them in an information management system developed with Flask. It also stores detections in a SQLite database, which continuously feeds into a newly automatically tagged dataset for future model updates, and is capable of displaying historical records of detections. The system's efficacy was evaluated across three key aspects: detection of suspicious objects, with an f1-score of 87.5%; detection of threatening postures, with 74%; and early detection, with 78%.The study's findings underscore the significance of processing complementary parts, facilitating the establishment of a more comprehensive contextual understanding of the scene. The quality and breadth of the training dataset are foundational for the success or failure of computer vision and machine learning models, particularly in the context of object detection, where a diverse array of images of varying sizes, shapes, and perspectives of the object to be detected is essential for the model to generalize correctly. In the case of body postures, support vector machine (SVM) models demonstrate efficacy, though they are constrained in their ability to establish spatio-temporal contexts.
Sistema remoto de detección de caídas en adultos mayores utilizando visión artificial
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2024-08)
Cuyo Gutiérrez, Jonathan Andrés
;
Castro Martin, Ana Pamela
This research project develops a monitoring system to detect falls in older adults using computer vision. According to the WHO, falls among older adults are a leading cause of injuries and can be fatal if not promptly addressed. The objective is to implement a system that identifies falls and notifies caregivers via Telegram. For fall detection, the YOLO v8s AI model trained on the COCO dataset, which includes over 220,000 images for key point and bounding box detection, has been selected. The system utilizes the Kinect V2, equipped with an RGB camera for live video, an infrared camera for nighttime operation, and a proximity sensor. The AI detects key points and bounding boxes in the images, while fall parameters are defined programmatically by analyzing aspect ratios and key point positions. In tests across three different scenarios, the system achieved a 92.92% accuracy in fall detection. Caregivers can access live monitoring through a web page, viewing real-time video. This system proves to be a valuable tool for elderly care, enabling a prompt and effective response to incidents, and significantly contributing to their well-being and quality of life.
Sistema de detección temprana de plagas en cultivos de mora mediante visión artificial
(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2024-08)
Arequipa Tipantuña, Jonathan Marcelo
;
Castro Martin, Ana Pamela
The blackberry production in Ecuador has experienced significant growth in recent years, becoming one of the most in-demand fruits. However, several factors, such as pests and diseases, affect the production and quality of this fruit, directly impacting the farmers' economy. Based on this context, the present research project focuses on the implementation of a pest and disease detection system using an artificial vision model, which serves as a tool to help farmers identify diseases in their crops quickly and efficiently. The system consists of a mobile application that integrates the artificial vision model. This model is designed to detect four types of pests with the highest incidence in the area. The application offers two detection modes: the first one through real-time video via the device's camera and the second one through image analysis, thus allowing the integration of multimedia resources for transmitting information to the model for analysis. The system has demonstrated an accuracy of 90.34% in pest detection, according to tests conducted in various environments where crops are located. The implementation of this system has significantly reduced pest identification errors by avoiding human errors and reducing economic losses by identifying pests and diseases at early stages of infection.

Ingeniería en Sistemas, Electrónica e Industrial

Examinar

Filtros

Ajustes

Ordenar por

resultados por página

Resultados de la búsqueda