"Artificial Intelligence-Based For Smart Monitoring And Health Analysis Of Blue Ternate Leaves "  Aaron L. Gulane...et al...  -  April 2024  - 72 pages.,;  illustrations  28cm. <br/><br/>Thesis <br/><br/> The general objective of this study is to develop an Artificial Intelligence-based health analysis and smart monitoring for Blue Ternate leaves. The research was conducted by the students taking the Bachelor of Science in Electronics Engineering at the University of Rizal System Morong-Rizal during the first and second semesters of the academic year 2023-2024. The location of the study will be at one of the residences of the researchers at Tatala, Binangonan, Rizal.  The researchers focused on implementing and evaluating smart technologies such as IoT sensors and image algorithms in blue ternate indoor cultivation. It aimed to provide and address the growing need for sustainable, technology-driven agriculture while preserving and promoting the cultivation of Blue Ternate plant. To determine the level of accuracy of the developed Artificial Intelligence-Based for Smart Monitoring and Health Analysis for Blue Ternate Leaves in terms of true positive, true negative, false positive, and false negative, the confusion matrix accuracy equation is used.  The researchers utilized descriptive and developmental research to assess the performance of sensors such as humidity, soil moisture, and temperature sensors and compare them to the image recognition of the device. They examined the effectiveness and accuracy of the A.I., using a record sheet to document the health of the plant. The record sheet is used to collect and retain data for evaluation responses from the plant regarding the accuracy of the technology, its potential benefits, and any concerns or suggestions they may have as it allows comprehensive responses. The researchers gathered various conditions of grown Clitoria Ternatea plants including healthy and unhealthy conditions of the plant and utilized resources such as humidity, soil moisture, temperature sensors, and Raspberry Pi to design and develop the device and test the system’s performance via IoT, which provided better conditions for the plants’ needs and will easily garner the needs of Clitoria Ternatea plants in terms of its condition. Sensors and A.I. imaging algorithms will enable the gathering of relevant information and transmit signals to the created website for the researchers to evaluate the device and establish its performance.  Researchers have designed an Artificial Intelligence-Based for Smart Monitoring and Health Analysis for Blue Ternate Leaves, employing Convolutional Neural Network (CNN) technology. This AI-powered device accurately assesses the health of Blue Ternate plants, distinguishing between healthy and unhealthy states with precision. By integrating Raspberry Pi, a computer screen, and a camera, the system efficiently evaluates plant health, which can detect the percentage of healthy and unhealthiness of the plant sample. The device demonstrated a remarkable 98% testing accuracy for the Confusion Matrix, ensuring reliable results. Furthermore, the device offers real-time monitoring of soil moisture, temperature, and humidity, enhancing the overall performance of smart monitoring for Blue Ternate plants.  In conclusion, the researchers have successfully developed an A.I.-based device for accurately assessing the health of Blue Ternate plants through image recognition, categorizing them as healthy or unhealthy. This system, supported by a database of collected images, provides precise evaluations of plant health on a dedicated website, showing efficient plant monitoring and management. Moreover, the A.I. system demonstrates remarkable accuracy in automatically checking the wellness of Blue Ternate plants using a camera, as evidenced by high testing accuracy revealed by the Confusion Matrix, distinguishing healthy and unhealthy plants with precision.  Additionally, the monitoring system enables real-time tracking of soil moisture, temperature, and humidity, with a dashboard offering graphical representations over various time intervals, ensuring effective and proper monitoring of plant health.  The following recommendations are hereby offered: Diversify by using other plants; future researchers should also specify particular illnesses for the A.I. to detect; they may assess the device's level of acceptability based on ISO/IEC 25010 criteria for the designed project. Additionally, it is advisable to consider utilizing unguided or unsupervised A.I. techniques to enhance the quality of data sets.