Ingeniería en Sistemas, Electrónica e Industrial
Permanent URI for this communityhttp://repositorio.uta.edu.ec/handle/123456789/1
Browse
5 results
Search Results
Item 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 PatricioThe 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.Item Sistema de monitoreo y control IOTpara cultivos agrícolas basado en la arquitectura Edge Cloud y Deep Learning(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2025-02) Amán Caiza Byron Alexander; Manzano Villafuerte Víctor SantiagoLow agricultural productivity is a significant challenge that impacts the efficiency and sustainability of the sector, especially in developing countries. In the case of hydroponic crops, this problem is aggravated by the lack of advanced monitoring and control systems. To address this limitation, the IoT Monitoring and Control System for Agricultural Crops based on Edge-Cloud and Deep Learning was developed, optimizing crop growth and management. The implementation of the IoT system for monitoring and control of agricultural crops based on Edge-Cloud and Deep Learning architectures allowed for real-time data processing, optimizing resource management and reducing manual intervention. It was developed in four layers. In the Edge Layer, devices collected pH and electrical conductivity data, while a camera captured the lettuce. In the Server Layer, data was processed and stored, the artificial intelligence model was trained and applied. In the Cloud Layer, a virtual network managed the information. Finally, in the Display Layer, an interface allowed real-time visualization, facilitating system monitoring. The results demonstrated the effectiveness of the system, reaching an accuracy of 99.76% and 98.80% in the measurement of pH and electrical conductivity. The lettuce disease recognition model achieved 87% in training and 84.92% in real tests, allowing early detection and reducing losses. The integration of IoT, Edge-Cloud and Deep Learning optimized monitoring and control, reducing costs and improving efficiency in the application of nutrients, guaranteeing a more sustainable system.Item Sistema de control de calidad de cultivo de fruta de temporada para etapa de precosecha empleando robótica aérea con planificación de trayectorias y visión artificial(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2024-08) Ashqui Balseca, Michelle Ivette; Aucatoma Matias, Bryan Paul; Córdova Córdova, Edgar PatricioCurrently, agriculture plays a fundamental role in the global economy. To meet the growing food demand, advanced technological tools have been integrated to optimize agricultural practices, known as Precision Agriculture (PA). These tools offer solutions that will mitigate the difficulties faced by farmers in their daily tasks. In this context, a study was carried out with the aim of implementing a quality control system for seasonal fruit crops for the pre-harvest stage using aerial robotics with trajectory planning and artificial vision. The importance of fruit quality control in Ecuador lies in its high demand both in the national and international markets. Technical standard NTE INEN 1872 establishes criteria for classifying apples according to quality grades for export and consumption. This system is based on the use of YOLOv8, a deep learning tool that evaluates the quality grade and classifies the different types of apples. The system consists of four stages: acquisition, processing, training, and visualization. In the acquisition stage, the Dji Tello drone was used to capture images or videos in real-time. The acquired data undergo preprocessing using OpenCV. A neural network was employed to train a model capable of accurately recognizing the type and quality grade of apples. For visualization, an intuitive graphical interface was designed to allow visual representation of the data derived from the trained model. The system algorithm was developed in Python due to its multiple libraries.Item Renderizado de coloración capilar utilizando inteligencia artificial(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Carrera de Telecomunicaciones, 2023-09) Balladares Armendariz, Johanna Elizabeth; Manzano Villafuerte, Víctor SantiagoThe importance in the world of hair coloring has become a trend and people who decide to change their hair color go to an esthetics where they usually ask for recommendations without knowing the final result. The purpose of creating a hair color rendering application is to provide a digital tool that shows color recommendations through a process of hair colorimetry by previewing a hair tone virtually. The application collects three physical characteristics: eye, skin and hair color that are stored in a MySQL database. The Support Vector Machine algorithm uses the TensorFlow and Keras libraries for the hair colorimetry process to predict hair color. The DeepLabv3 model segments the hair of a loaded image resulting in a mask. Finally, a process that couples the results of the previous algorithms is applied to display the customized hair color using the RGB model. The evaluation of the hair colorimetry algorithm uses cross-validation and obtains a value of 0.78 in training and testing, being able to predict the color adequately. The quality metrics for the DeepLabv3 model present average results: recall 98%. Dice Coefficient 98% and IoU 97%, considering that the model has good ability to segment hair in images. The application offers benefits of personalized and realistic counseling, providing a favorable change in the user's appearance by combining theoretical and technological creativity.Item Controlador para alimentación de peces empleando Deep Learning(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Electrónica y Automatización, 2022) Cañar Yumbolema, Willam Patricio; Galarza Zambrano, Eddie EgbertoThis degree project consists of the design and implementation of a controller for fish feeding using Deep Learning for which a Raspberry minicomputer and a lowcost Web camera were used, this significantly reduced the investment for project development. The main objective consists in the creation of a manually labeled dataset (set of images) of several zebra-type fish (Danio Rerio) located inside a fish tank uniformly illuminated by white LED light. In this case, it was decided to use two videos with the presence of 4 and 6 fish, respectively. Through the use of a computational algorithm, the sequence of images of the fish where their movements can be identified were obtained. This information is used to train a convolutional network using the ACF (Aggregated Channel Characteristics) image object detection algorithm. Once the location of the fish inside the fish tank is determined, the feeling of the school is identified through the implementation of three zones, that is, the developed algorithm will allow knowing if the fish are in a satisfaction zone, a normal zone or a normal zone. feeding. Finally, the FuzzySN, FuzzySH and FuzzyNH indices contain the feeling of the fish and are the inputs of a fuzzy controller which in turn contains the feeding rules based on the natural behavior of the fish; In this way, the developed system is capable of feeding the fish automatically. The minimum identification error reached was 29.5%, but the identification of the behavior of the school of fish had a success rate of 100%. The test carried out to validate the algorithm was given for a case of manual feeding by an operator, where the system was able to correctly identify the feeling of the school as satisfied.