Classification of Oil Palm Fresh Fruit Bunches Based on Their Maturity Using Thermal Imaging Technique

Abstract

The maturity of oil palm Fresh Fruit Bunches (FFB) is considered to be a significant factor that affects the profitability and salability of palm oil FFB. Typical methods of grading FFB consist of physical grading of fresh fruit, which is time-consuming and expensive, and the results are prone to human error. Therefore, this research attempts to formulate a thermal imaging method to indicate the precise maturity of oil palm fruits. A total of 297 oil palm FFBs were collected. The samples were divided into three groups: under-ripe, ripe, and over-ripe. Afterward, all the samples were scanned using a thermal imaging camera to calculate the real temperature of each sample. In order to normalize the measurement, the difference between the average temperature of the palm bunch and the ambient temperature (∆Temp) was considered as the main parameter. The results indicated that the mean ∆Temp of oil palm FFBs decreased consistently from under-ripe to over-ripe. The results of the ANOVA test demonstrated that the observed significance value was less than 0.05 in terms of ∆Temp, so there is a statistically significant difference in the means of all three maturity categories. It can be concluded that ∆Temp is a reliable index to classify the FFBs of oil palm. The classification analysis was conducted using the ∆Temp of the FFBs and its application as an index in Linear Discriminant Analysis (LDA), Mahalanobis Discriminant Analysis (MDA), Artificial Neural Network (ANN), and Kernel Nearest Neighbor (KNN). The highest degrees of overall accuracy (99.1% and 92.5%) were obtained through the ANN method. This study concludes that thermal images can be used as an index of oil palm maturity classification.

Annotated Datasets of Oil Palm Fruit Bunch Piles for Ripeness Grading Using Deep Learning

“…Bhd (FASSB) 4 4 classes (unripe, under-ripe, ripe and overripe. 80 FFBs were 208 (images) palm oil estate in Johor 5 2 classes (ripe and unripe) 264 (images) Local dataset from Indonesia 6 7 classes (Ripening, Raw, less Ripped, Almost Ripped, Ripped, Perfectly ripped, too ripped) 400 (images) Local dataset from Malaysia 8 4 classes (unripe, under-ripe, ripe and overripe) 120 (images) United Plantation Research and Development in Teluk Intan (UPRD), Perak, Malaysia 10 3 categories (under-ripe, ripe, and over-ripe) 297 oil palm FFBs (images) Local dataset from Malaysia 11 2 categories (ripe, unripe) 490 FFBs (images) Local dataset from Malaysia 13 3 categories (under-ripe, ripe and overripe) 120 oil palm FFBs (images) Local dataset from West Java 15 3 categories (under-ripe, ripe and overripe) 180 FFBs (images) …”

“…Unfortunately, the open datasets used in the current study on oil palm ripeness are not available. Comparatively, the current research’s typical dataset attempts to increase the grade and output of refined palm oil 2 – 6 , 8 , 10 , 11 , 13 , 15 . The video dataset employed in this study, however, is concentrated on offering an assessment of the oil palm FFB maturity level, especially in oil palm processing facilities.…”

Pulsed Thermography as a Reliable Tool to Detect Presymptomatic Stages of Botrytis cinerea Infection in Plants

“…3,4 In this scenario, the advent of fast and noninvasive high-throughput analyses based on infrared thermography (IT) represents an outstanding innovation in agriculture, with multiple applications, such as irrigation scheduling aimed to water saving, 5−7 predicting crop yields, 8 detecting foreign substances in plants and food, 9,10 and assessing fruit maturity. 11 A growing body of experimental evidence has demonstrated that IT techniques can also be efficiently used to monitor the pathological status of plants and predict the occurrence of plant diseases by analyzing the onset of symptoms at an early stage. 12− 15 A valuable indicator of plant health status is the leaf temperature, which may vary in response to both biotic and abiotic stress factors.…”

“…The advancement of automatic detection tools to picture plant and crop diseases is one of the main goals of precision agriculture. In particular, the use of smart sensors combined with artificial intelligence resources has paved the way for real-time analyses of plants’ physiological status and their fundamental interaction with the environment. , In this scenario, the advent of fast and noninvasive high-throughput analyses based on infrared thermography (IT) represents an outstanding innovation in agriculture, with multiple applications, such as irrigation scheduling aimed to water saving, − predicting crop yields, detecting foreign substances in plants and food, , and assessing fruit maturity . A growing body of experimental evidence has demonstrated that IT techniques can also be efficiently used to monitor the pathological status of plants and predict the occurrence of plant diseases by analyzing the onset of symptoms at an early stage. − …”

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

“…To aid the process, technologies have been developed over years to determine the maturity level of FFB. 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.…”