Lettuce life stage classification from texture attributes using machine learning estimators and feature selection processes

Abstract

Classification of lettuce life or growth stages is an effective tool for measuring the performance of an aquaponics system. It determines the balance in water nutrients, adequate temperature and lighting, other environmental factors, and the system’s productivity to sustain cultivars. This paper proposes a classification of lettuce life stages planted in an aquaponics system. The classification was done using the texture features of the leaves derived from machine vision algorithms. The attributes underwent three different feature selection processes, namely: Univariate Selection (US), Recursive Feature Elimination (RFE), and Feature Importance (FI) to determine the four most significant features from the original eight attributes. The features selected were used for training four estimators from Decision Trees Classifier (DTC), Gaussian Naïve Bayes (GNB), Stochastic Gradient Descent (SGD), and Linear Discriminant Analysis (LDA). The models trained using DTC and SGD were then optimized as they have hyperparameters for tuning. A comparative analysis among Machine Learning (ML) algorithms was conducted to identify the best-performing model with the given application. The best features were derived from US and FI as they have the same top four features using the DTC estimator optimized with the hyperparameters tuned to max depth having 5, criterion equated to ‘Gini', and splitter was set to 'Best'. The accuracy obtained from cross-validation evaluation resulted in 87.92%. Considering consistency with hold-out validation, LDA outperforms optimized DTC even with lower accuracy of 86.67%. This accuracy of LDA outperformed DTC due to its sufficient fit for generalizing the testing data on classifying lettuce growth stage.

Bioinspired Optimization of Germination Nutrients Based on Lactuca sativa Seedling Root Traits as Influenced by Seed Stratification, Fortification and Light Spectrums

“…The computational root phenotype model consists of processes of root segmentation and root numerical traits extraction. Graph-cut segmentation is a computer vision technique based on energy minimization that computationally determines the arrangement of atoms in physical space or picture elements for images in terms of color features through K-means clustering (Concepcion, Lauguico, Alejandrino, Dadios, & Sybingco, 2020;Lauguico, Concepcion, Alejandrino, Tobias, & Dadios, 2020). It uses lazy snapping that snaps true object pixels from low contrast edges (Concepcion, Lauguico, Alejandrino, Dadios, & Sybingco, 2020).…”

Machine Learning Methods for Early-Stage Diagnosis of Parkinson's Disease Through Handwriting Data

“…These were Logistic Regression, Multilayer Perceptron, and Naive Bayes. Numerous research has shown that these algorithms are able to create accurate models, some even outperform other highly sophisticated classification methods [25,26,27,29]. As such, these three algorithms were chosen for their simple structure, functionality, as well as their availability in the scikit learn software.…”

“…As of writing, sklearn can only implement univariate filter selection methods and the recursive feature elimination algorithm. As such, Univariate Selection (US) and Recursive Feature Elimination (RFE) were done as implemented by [29].…”

“…It could be seen that under Univariate Selection, descending ranking of the features were as follows: 'STD_HT', 'MIN_BETWEEN_ET_HT', 'RMS', 'MAX_BETWEEN_ET_HT', 'AGE', 'MRT', 'GENDER', 'MIN_HT', 'MAX_HT', 'STD_DEVIATION_ET_HT', 'RIGHT/LEFT-HANDED', and 'CHANGES_FROM_NEGATIVE_TO_ POSITIVE_BETWEEN_ET_HT'. As done by [29], features with 0 k scores were dropped. The remaining nine features were then selected and used to build three models.…”

Non-destructive in Situ Measurement of Aquaponic Lettuce Leaf Photosynthetic Pigments and Nutrient Concentration Using Hybrid Genetic Programming

“…For each row of combined 18 predictors, the Hampel algorithm computes the median and standard deviation of the 10 samples from each side, and if the original sample is more than thrice the computed standard deviation, it will be replaced by the computed median. Feature-based machine learning is a subset of computational intelligence that trains a predictive model, whether it is for classification or regression, using multiple categorized numerical data as its features (Lauguico, Concepcion II, Alejandrino, Tobias, & Dadios, 2020;Loresco & Dadios, 2018). The features tested in the study were generalized processing regression (GPR), regression-based support vector machine (RSVM), regression tree (RTree), adaptive neuro-fuzzy inference system (ANFIS), and multigene symbolic regression genetic programming (MSRGP).…”