A Comparative Analysis of Machine Learning Algorithms Modeled from Machine Vision-Based Lettuce Growth Stage Classification in Smart Aquaponics

Abstract

The arising problem on food scarcity drives the innovation of urban farming. One of the methods in urban farming is the smart aquaponics. However, for a smart aquaponics to yield crops successfully, it needs intensive monitoring, control, and automation. An efficient way of implementing this is the utilization of vision systems and machine learning algorithms to optimize the capabilities of the farming technique. To realize this, a comparative analysis of three machine learning estimators: Logistic Regression (LR), K-Nearest Neighbor (KNN), and Linear Support Vector Machine (L-SVM) was conducted. This was done by modeling each algorithm from the machine vision-feature extracted images of lettuce which were raised in a smart aquaponics setup. Each of the model was optimized to increase cross and hold-out validations. The results showed that KNN having the tuned hyperparameters of n_neighbors=24, weights='distance', algorithm='auto', leaf_size = 10 was the most effective model for the given dataset, yielding a cross-validation mean accuracy of 87.06% and a classification accuracy of 91.67%.

Differential Effects of Potassium Chloride on Vascular Tissues, Morphological Traits and Germination of Tomato with Sperm Swarm-based Nutrient Optimization

“…Genetic programming, a computational intelligence technique that is based on Charles Darwin's evolutionary concept [39], through MATLAB R2021b GPTIPS v2 tool was employed to develop the fitness model for the five phytomorphological phenotypes as characterized by the potassium chloride concentration ([KCl], mM) and cultivation period (t, DAS) Eq. ( 4) [40,41]. The multigene symbolic regression genetic programming (MSRGP) hyperparameters were configured as 50 population size, 100 maximum generations, tournament size of 50, 0.05 elite fraction with enabled lexicographic selection pressure and 0.1 Pareto tournament probability, 10 maximum genes, 5 maximum tree depth, 0.1 ephemeral random constant, crossover rate of 0.85, the mutation rate of 0.14 and mathematical function set of {times, minus, plus, sqrt, square, sin, cos, log, cube, neg, abs}.…”

Cultivating Tomorrow: A Comprehensive Review of Agricultural Innovations in the Philippines (2018-2023)

“…Agricultural innovations cover a diverse array of technologies, techniques, systems, and smart farming practices, incorporating cutting-edge tools such as AI and IoT [1,2,3], advanced sensors [4,5], robotics [6], and data analytics [7,8]. These innovations, as demonstrated by studies [9,10], are designed to tackle challenges in agriculture, aiming to enhance e ciency, sustainability, productivity, and decision-making processes.…”

Transforming Philippine Agriculture Through Data-driven Innovation: A Quantitative Landscape Assessment to Prioritize Technological Solutions

“…Agricultural innovations refer to the range of technologies, techniques, systems, and smart farming practices that harness cutting-edge tools like AI, and IoT (Espineli and Lewis, 2021; Tagle et al, 2018;Arago et al, 2022), advanced sensors (Cruz et al, 2018;Bacsa et al, 2019), robotics (De Padua et al, 2021) and data analytics (Lauguico et al, 2020: Velasco, 2020 to solve various e ciency, sustainability, productivity, and decision-making challenges in agriculture (Dhanaraju et al, 2022;Abashidze, 2023).…”