Deep Learning to Detect and Classify the Purity Level of Luwak Coffee Green Beans

Hendrawan, Yusuf; Widyaningtyas, Shinta; Fauzy, Muchammad Riza; Sucipto, Sucipto; Damayanti, Retno; Riza, Dimas Firmanda Al; Hermanto, Mochamad Bagus; Sandra, Sandra

doi:10.47836/pjst.30.1.01

Cited by 11 publications

(4 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The model with the highest validation accuracy was then tested for its performance using data testing. The data testing results were displayed using a confusion matrix [45] [46].…”

Section: Methodsmentioning

confidence: 99%

Convolutional Neural Network to Detect the Optimal Water Content of Cassava Chips During the Drying Process

Hendrawan¹,

Rohmatulloh²,

Ilmi³

et al. 2022

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

Cassava chips are used as raw materials to manufacture modified cassava flour. To produce high-quality modified cassava flour, a drying process for cassava chips is required to produce optimal water content in the range of 15-18% wb. This study aims to detect the optimal water content of cassava chips during the drying process in a hybrid hot-air tray dryer with computer vision using a convolutional neural network. Three categories of cassava chips' water content during the drying process are wet (water content of 55-70% wb), semi-dry (20-40% wb), and optimal dry (15-18% wb). In this study, the performance of four types of the pre-trained convolutional neural network, i.e., AlexNet, GoogLeNet, ResNet-50, and SqueezeNet, were compared by using different optimizers (SGDm, Adam, and RMSProp) and different learning rate values, 0.00005 and 0.0001, resulting in 24 types of experimental design. The results showed 12 convolutional neural network models with perfect validation accuracy. AlexNet with the SGDm optimizer and learning rate of 0.00005 was determined as the best model because of its stable training iteration process that experienced no fluctuations, perfect validation accuracy, specifically 100%, as well as perfect testing accuracy was 100%, and fastest training and validation process time, notably 32 minutes. This best convolutional neural network model will later be used to develop a rapid, realtime, and accurate hybrid hot-air tray dryer with computer vision to maintain cassava chip products with optimal water content.

show abstract

“…The model with the highest validation accuracy was then tested for its performance using data testing. The data testing results were displayed using a confusion matrix [45] [46].…”

Section: Methodsmentioning

confidence: 99%

Convolutional Neural Network to Detect the Optimal Water Content of Cassava Chips During the Drying Process

Hendrawan¹,

Rohmatulloh²,

Ilmi³

et al. 2022

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

show abstract

“…Hendrawan et al [20] and colleagues suggested utilizing Deep Learning approaches to detect and categorize purity levels in luwak coffee green beans. The study sought to detect and classify the purity of Luwak coffee green beans into four categories: deficient (0-25%), low (25-50%), medium (50-75%), and high (75-100%).…”

Section: Related Workmentioning

confidence: 99%

Using Hierarchical Agglomerative Clustering in E-Nose for Coffee Aroma Profiling: Identification, Quantification, and Disease Detection

Selvanarayanan,

Rajandran,

Alotaibi

2023

I2M

View full text Add to dashboard Cite

Numerous coffee devotees believe that the coffee smell plays a vital role in the coffeedrinking insight, complementing the taste and enhancing delight. In the traditional strategy, aroma patterns and profiles are observed by extensive investigation of human olfaction. However, the outcome tends to be imprecise. Tackling the difficulties encountered in distinct scent profiles linked to various coffee bean varieties, including Arabica, Robusta, Monsoon Malabar, Chikmagalur, and Coorg coffee, as well as diverse roasting techniques, through the utilization of Electronic Nose Applications for the investigation of coffee aromas. The suggested methodology employs e-nose technology utilizing conducting polymer sensors to detect aroma volatile chemicals found in coffee, including furaneol, 2-methylisoborneol, and 3-methylindole. The e-nose olfactory characteristics of coffee beans at various stages of roasting are systematically examined and discernible patterns are duly identified. The average intensity of the coffee aroma perceived at a distance of 10 centimeters was rated as 3.9 on a scale of 5. The observed standard deviation of coffee aroma intensity at a distance of 10 centimeters was determined to be 3.8 on a scale of 5. The p-value associated with the disparity in average fragrance scores was determined to be 0.05.

show abstract

“…Penambangan data dalam pembelajaran mesin memainkan peran besar dalam hal data eksperimen dan perhitungan kompleks untuk interpretasi [18] [19]. Penggabungan machine learning dalam pengembangan pengetahuan masa depan melalui interpretasi data otomatis dengan bantuan berbagai model seperti klasifikasi, regresi, pengelompokan, dan lainnya [20] [21].…”

Section: Pendahuluanunclassified

SmartAgro-Spectral: Teknik Pengukuran Kandungan Nitrit Pada Sarang Burung Walet Berbasis Spektral Menggunakan Metode Regresi Linier

Hudaya,

Hepi Ludiyati,

Feni Isdaryani

et al. 2023

J. Appl. Comput. Sci. Technol.

View full text Add to dashboard Cite

Indonesia is the largest exporter of edible bird’s nest (EBN) to China, involving many EBN farmers from various regions in Indonesia. Therefore, a portable device is needed to rapidly and accurately measure the required quality of SBW to avoid rejection by Chinese buyers, which could result in significant losses. Consequently, for this purpose, an electronic instrument has been developed. smartAgro-Spectral is a microcontroller-based electronic instrument that measures nitrite content in edible bird’s nest (EBN) using linear regression method in machine learning calculations. This instrument can measure nitrite content based on the intensity of colors produced by EBN products. The coloring process is carried out by mixing EBW powder with Sulphanilamide solution and N-(1-naphthyl) Ethylenediamine Dihydrochloride (NED) solution. The concentration of EBN solution is normalized to values between 0.2 ppm and 0.7 ppm. The measurement process is carried out by emitting 18 waves of the light spectrum. The intensity of the 18 wavelengths of the measured light spectrum was selected based on the strong correlation between the intensity of the light spectrum and the value of nitrite content in the EBN product. The measurement results show that the intensity of the light spectrum that has a strong linear correlation is at wavelengths of 460 nm, 485 nm, 510 nm, 535 nm, and 610 nm. So, smartAgro-Spectral electronic instruments can be realized based on the intensity relationship of each wavelength through multiple linear regression analysis, and are able to linearly measure nitrite content in EBW products with a precision level of 99.85% and an accuracy rate of 99.85%.

show abstract

Deep Learning to Detect and Classify the Purity Level of Luwak Coffee Green Beans

Cited by 11 publications

References 54 publications

Convolutional Neural Network to Detect the Optimal Water Content of Cassava Chips During the Drying Process

Convolutional Neural Network to Detect the Optimal Water Content of Cassava Chips During the Drying Process

Using Hierarchical Agglomerative Clustering in E-Nose for Coffee Aroma Profiling: Identification, Quantification, and Disease Detection

SmartAgro-Spectral: Teknik Pengukuran Kandungan Nitrit Pada Sarang Burung Walet Berbasis Spektral Menggunakan Metode Regresi Linier

Contact Info

Product

Resources

About