Oil Palm Fruits Ripeness Classification Based on the Characteristics of Protein, Lipid, Carotene, and Guanine/Cytosine from the Raman Spectra

Abstract

The capacity of palm oil production is directly affected by the ripeness of the fresh fruit bunches (FFB) upon harvesting. Conventional harvesting standards rely on rigid harvesting scheduling as well as the number of fruitlets that have loosened from the bunch. Harvesting is usually done every 10 to 14 days, and an FFB is deemed ready to be harvested if there are around 5 to 10 empty sockets on the fruit bunch. Technology aided by imaging techniques relies heavily on the color of the fruit bunch, which is highly dependent on the surrounding light intensities. In this study, Raman spectroscopy is used for ripeness classification of oil palm fruits, based on the molecular assignments extracted from the Raman bands between 1240 cm−1 and 1360 cm−1. The Raman spectra of 52 oil palm fruit samples which contain the fingerprints of different organic compounds were collected. Signal processing was applied to perform baseline correction and to reduce background noises. Characteristic data of the organic compounds were extracted through deconvolution and curve fitting processes. Subsequently, a correlation study between organic compounds was developed and eight hidden Raman peaks including protein, beta carotene, carotene, lipid, guanine/cytosine, chlorophyll-a, and tryptophan were successfully located. Through ANOVA statistical analysis, a total of six peak intensities from proteins through Amide III (β-sheet), beta-carotene, carotene, lipid, guanine/cytosine, and carotene and one peak location from lipid were found to be significant. An automated oil palm fruit ripeness classification system deployed with artificial neural network (ANN) using the seven signification features showed an overall performance of 97.9% accuracy. An efficient and accurate ripeness classification model which uses seven significant Raman peak features from the correlation analysis between organic compounds was successfully developed.

Oil Palm Bunch Ripeness Classification and Plantation Verification Platform: Leveraging Deep Learning and Geospatial Analysis and Visualization

“…A comprehensive review of literature and research pertaining to the utilization of machine learning and deep learning in the classification and assessment of oil palm fruit or bunch ripeness revealed the adoption of both classical machine learning [26][27][28][29][30] and deep learning [17] techniques. The findings from this investigation indicate that deep learning methodologies generally outperform other machine learning algorithms in accurately classifying the ripeness of oil palm bunches.…”

“…Previous research has proposed various classical machine learning methods, such as Artificial Neural Networks (ANN), K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Naïve Bayes, Regression, Decision Tree (DT), and Fuzzy Logic, for assessing the ripeness of oil palm fruit bunches or fruits, in comparison with Convolutional Neural Networks (CNN). Although studies have highlighted CNN's superior classification accuracy [26], many prior investigations utilizing machine learning algorithms like KNN, SVM, and ANN have achieved remarkably high accuracy rates in classifying oil palm fruits or bunch ripeness, ranging between 97% and 100% [27][28][29][30]. Regarding the application of deep learning in classifying and detecting the ripeness of bunches or oil palm fruits, the analysis reveals a classification accuracy stratified into the following three levels: very accurate (accuracy exceeding 95%) [24,25], highly accurate (accuracy ranging between 80% and 94%), and moderately accurate (accuracy between 60% and 79%) [22,23], with the majority of outcomes falling within the highly accurate range.…”

“…A comprehensive review of the related literature and research reveals the utilization of machine learning techniques for processing images to determine the ripeness levels of harvested oil palm fruits, categorized into multiple levels to aid in sorting and assessing the quality of factory purchases for setting purchase prices [17][18][19][20][21][22][23][24][25][26][27][28][29][30]. Notably, the evaluation of classification accuracy demonstrates the high precision achieved through machine learning, particularly with the application of deep learning algorithms [18][19][20][21][22][23][24][25][26].…”

“…However, despite these advancements, the analysis underscores several limitations in existing research. Notably, the absence of an application or platform capable of real-time data processing [17][18][19][20][21][22][23][24][25][26], reliance on a limited number of datasets for model creation leading to potential overfitting issues in practical scenarios [17,[21][22][23]28,29], and the lack of clarity regarding the number of datasets used for modeling and testing [24,25]. Furthermore, the digital images employed for classification predominantly feature harvested oil palm fruit with intact backgrounds, rendering them unsuitable for on-tree oil palm fruit classification [17][18][19][20][21][22][23][24][25][26].…”

Enabling chlorophyll photo-response for in-line real-time noninvasive direct probing of the quality of palm-oil during mill process

“…These include computer vision (using red-green-blue imaging) integrated with laser-light backscattering techniques 6 , spectroscopy 7 , hyperspectral imaging 8 , thermal imaging 9 , inductive frequency technique 10 and electrochemical sensor based on the fruit battery principle 11 . More recently, Raman 7,12,13 , Infrared 14 , NIR 15,16 , optical sensor 17 or diffuse reflectance spectroscopy 18 have also been used to classify oil palm fresh fruit maturity based on carotene and chlorophyll content. Carotene and chlorophyll are extracted along with palm oil during the extraction process from the palm fruit mesocarp.…”