scholarly journals Estimation of percentage of impurities in coffee using a computer vision system

Author(s):  
Anderson G. Costa ◽  
Eudócio R. O. da Silva ◽  
Murilo M. de Barros ◽  
Jonatthan A. Fagundes

ABSTRACT The quality and price of coffee drinks can be affected by contamination with impurities during roasting and grinding. Methods that enable quality control of marketed products are important to meet the standards required by consumers and the industry. The purpose of this study was to estimate the percentage of impurities contained in coffee using textural and colorimetric descriptors obtained from digital images. Arabica coffee beans (Coffea arabica L.) at 100% purity were subjected to roasting and grinding processes, and the initially pure ground coffee was gradually contaminated with impurities. Digital images were collected from coffee samples with 0, 10, 30, 50, and 70% impurities. From the images, textural descriptors of the histograms (mean, standard deviation, entropy, uniformity, and third moment) and colorimetric descriptors (RGB color space and HSI color space) were obtained. The principal component regression (PCR) method was applied to the data group of textural and colorimetric descriptors for the development of linear models to estimate coffee impurities. The selected models for the textural descriptors data group and the colorimetric descriptors data group were composed of two and three principal components, respectively. The model from the colorimetric descriptors showed a greater capacity to estimate the percentage of impurities in coffee when compared to the model from the textural descriptors.

Author(s):  
Hervé Cardot ◽  
Pascal Sarda

This article presents a selected bibliography on functional linear regression (FLR) and highlights the key contributions from both applied and theoretical points of view. It first defines FLR in the case of a scalar response and shows how its modelization can also be extended to the case of a functional response. It then considers two kinds of estimation procedures for this slope parameter: projection-based estimators in which regularization is performed through dimension reduction, such as functional principal component regression, and penalized least squares estimators that take into account a penalized least squares minimization problem. The article proceeds by discussing the main asymptotic properties separating results on mean square prediction error and results on L2 estimation error. It also describes some related models, including generalized functional linear models and FLR on quantiles, and concludes with a complementary bibliography and some open problems.


Author(s):  
Ati Atul Quddus

Abstrak Penelitian ini bertujuan untuk menduga kandungan energi bruto tepung ikan untuk bahan pakan ternak menggunakan teknologi Near Infrared (NIR). Tepung ikan yang digunakan dalam penelitian ini diperoleh dari poultry shop yang ada di beberapa daerah di Indonesia dan industri pakan ternak. Penelitian ini menggunakan 50 tepung ikan. Tiga puluh lima sampel digunakan untuk kalibrasi, sedangkan 15 sampel digunakan untuk validasi. Pengukuran NIR reflektan menggunakan sistem NIR. Energi bruto diukur menggunakan bomb calorimeter. Data dianalisis dengan menggunakan regresi linier berganda (RLB) dan Principal Component Regression (PCR). Persamaan kalibrasi dari reflektan dianalisis menggunakan 29 panjang gelombang untuk memprediksi energi bruto. Hasil dari validasi menunjukkan akurasi yang tinggi dengan standar eror dan koefisien variasi untuk energi bruto yaitu 6,6 Kkal/Kg dan 0,2%. Persamaan kalibrasi dari metode PCR menggunakan data absorban. Hasil dari validasinya menunjukkan kurang akurasi dengan nilai standar eror dan koefisien variasi yaitu 119,2 Kkal/kg dan 4,16%. Kata kunci : energi bruto, NIR, RLB, PCR Abstract This experiment was aimed to predict gross energy (GE) content of fishmeal by using Near Infrared (NIR) technology. Fishmeal that was used in this experiment was obtained from the poultry shop in several regions in Indonesia and from animal feed industries. This experiment was conducted by using 50 fishmeals. Thirty five samples out of 50 samples fishmeal was used to develop the NIR of calibration and the rest 15 samples was used to test the accuracy of the calibration. NIR reflectant was measured by NIR system. Gross energy was measured by bomb calorimeter. Collected data were analyzed by using multivariate linier regression (MLR) and principal component regression (PCR). Calibration equation of reflectant was analyzed by using 29 wavelengths for predicting GE. The results of the validation indicated high accuracy with standard error and coefficient of variation for GE: SEp = 6.6 Kkal/Kg, CV = 0.2 % . Calibration equation was obtained from PCR method by using absorbent data. The result of the validation indicated less accuracy with standard error and coefficient of variation for GE: SEp = 119.92 Kkal/Kg, CV = 4.16% . Keywords : Gross Energy, Near infrared Reflectant (NIR), fishmeal, Multivariate Linier Regression (MLR), Principal Component Regression (PCR)


Foods ◽  
2020 ◽  
Vol 9 (2) ◽  
pp. 113 ◽  
Author(s):  
Razieh Pourdarbani ◽  
Sajad Sabzi ◽  
Davood Kalantari ◽  
José Luis Hernández-Hernández ◽  
Juan Ignacio Arribas

Since different varieties of crops have specific applications, it is therefore important to properly identify each cultivar, in order to avoid fake varieties being sold as genuine, i.e., fraud. Despite that properly trained human experts might accurately identify and classify crop varieties, computer vision systems are needed since conditions such as fatigue, reproducibility, and so on, can influence the expert’s judgment and assessment. Chickpea (Cicer arietinum L.) is an important legume at the world-level and has several varieties. Three chickpea varieties with a rather similar visual appearance were studied here: Adel, Arman, and Azad chickpeas. The purpose of this paper is to present a computer vision system for the automatic classification of those chickpea varieties. First, segmentation was performed using an Hue Saturation Intensity (HSI) color space threshold. Next, color and textural (from the gray level co-occurrence matrix, GLCM) properties (features) were extracted from the chickpea sample images. Then, using the hybrid artificial neural network-cultural algorithm (ANN-CA), the sub-optimal combination of the five most effective properties (mean of the RGB color space components, mean of the HSI color space components, entropy of GLCM matrix at 90°, standard deviation of GLCM matrix at 0°, and mean third component in YCbCr color space) were selected as discriminant features. Finally, an ANN-PSO/ACO/HS majority voting (MV) ensemble methodology merging three different classifier outputs, namely the hybrid artificial neural network-particle swarm optimization (ANN-PSO), hybrid artificial neural network-ant colony optimization (ANN-ACO), and hybrid artificial neural network-harmonic search (ANN-HS), was used. Results showed that the ensemble ANN-PSO/ACO/HS-MV classifier approach reached an average classification accuracy of 99.10 ± 0.75% over the test set, after averaging 1000 random iterations.


Energies ◽  
2021 ◽  
Vol 14 (20) ◽  
pp. 6662
Author(s):  
Da Zhang ◽  
Shuailin Chen

To implement the non-contact detection of contamination on insulators, a contamination severity assessment methodology using the deep learning of the colored image information of insulators can be used. For the insulator images taken at the substation site, a mathematical morphology-improved optimal entropic threshold (OET) method is utilized to extract the insulator from the background. By performing feature calculations of insulator images in RGB and HSI color spaces, sixty-six color features are obtained. By fusing the features of the two color spaces using kernel principal component analysis (KPCA), fused features are obtained. The recognition of contamination grades is then accomplished with a deep belief network (DBN) that consists of a three-layered restricted Boltzmann machine. The experimental results of the images taken on-site show that the fused features obtained by the KPCA can fully reflect the contamination state of the insulators. Compared with the identification obtained using RGB or HSI color-space features alone, accuracy is significantly improved, and insulator contamination grades can be effectively identified. The research provides a new method for the accurate, efficient, and non-contact detection of insulator contamination grades.


Geosciences ◽  
2019 ◽  
Vol 9 (10) ◽  
pp. 413 ◽  
Author(s):  
Logan Clark ◽  
Ryan Fogt

The relationship between Southern Hemisphere middle and high-latitude regions has made it possible to generate observationally-based Antarctic pressure reconstructions throughout the 20th century, even though routinely collected observations for this continent only began around 1957. While nearly all reconstructions inherently assume stability in these relationships through time and in the absence of direct observations, this stationarity constraint can be fully tested in a model setting. Seasonal pressure reconstructions based on the principal component regression (PCR) method spanning 1905–2013 are done entirely within the framework of the Community Atmospheric version 5 (CAM5) model in this study in order to evaluate this assumption, test the robustness of the PCR procedure for Antarctic pressure reconstructions and to evaluate the CAM5 model. Notably, the CAM5 reconstructions outperformed the observationally-based reconstruction in every season except the austral summer. Other tests indicate that relationships between Antarctic pressure and pressure across the Southern Hemisphere remain stable throughout the 20th century in CAM5. In contrast, 20th century reanalyses all display marked changes in mid-to-high latitude pressure relationships in the early 20th century. Overall, comparisons indicate both the CAM5 model and the pressure reconstructions evaluated here are reliable estimates of Antarctic pressure throughout the 20th century, with the largest differences between the two resulting from differences in the underlying reconstruction predictor networks and not from changes in the model experiments.


2020 ◽  
Vol 4 (4) ◽  
pp. 502-511
Author(s):  
Mardiantono Mardiantono ◽  
Fachruddin Fachruddin ◽  
Zulfahrizal Zulfahrizal

Abtrak. Kadar Air merupakan salah satu komponen penting dalam beras ketan putih yang dapat mempengaruhi kualitas dari beras ketan putih. Penelitian ini bertujuan menguji dan mengevaluasi teknologi NIRS sebagai metode cepat dan tepat dalam memprediksi kadar air beras ketan dengan metode Principal Component Regression (PCR) serta menentukan metode koreksi spektrum yang terbaik dan akurat untuk memprediksi kadar air beras ketan dengan menggunakan pretreatment De- Trending, Derivative-2, dan Standart Normal Variate (SNV). Penelitian ini menggunakan beras ketan putih yang didapat dari pasar Rukoh Banda Aceh, yang berjumlah 35 sampel. Perlakuan yang diberikan adalah tanpa perendaman, dibasahi, dan perendaman selama 5, 10, 15, 20, dan 25 menit. Prediksi kadar air beras ketan dengan NIRS menggunakan alat self developed FT-IR IPTEK T-1516 dan metode referensi yang digunakan adalah metode gravimetri yang berdasarkan pada Association of Official Analytical Chemists (AOAC). Pengolahan data menggunakan Unsclambers sofware® X version 10.5. Hasil penelitian menunjukkan bahwa NIRS dengan metode PCR mampu menghasilkan model yang baik untuk pendugaan beras ketan. Penelitian ini menghasilkan empat model pendugaan kadar air beras ketan dimana satu model tergolong very good performance (RPD3) dan tiga model tergolong good model performance (RPD2) sehingga dapat dikatakan bahwa semua model yang dihasilkan layak dan baik untuk pendugaan kadar air beras ketan. Pretreatment terbaik pada penelitian ini adalah Standart Normal Variate (SNV) dengan nilai RPD 3,12, r sebesar 0,95, R2 sebesar 0,89, dan RMSEC sebesar 2,34.Estimation of White Gluttony Rice Rate With NIRS Technology Using Principal Component Regression Method (Pretreatment De-Trending, Derivative-2, dan Standart Normal Variate)Abstract. Water content is one important component in white glutinous rice which can affect the quality of white glutinous rice. This study aims to test and evaluate NIRS technology as a fast and precise method for predicting glutinous rice water content with the Principal Component Regression (PCR) method and determine the best and accurate spectrum correction method for predicting glutinous rice water content using the De-Trending, Derivative pretreatment -2, and Standard Normal Variate (SNV). This study uses white sticky rice obtained from the Rukoh market in Banda Aceh, which amounted to 35 samples. The treatment given is without soaking, soaking, and soaking for 5, 10, 15, 20, and 25 minutes. The prediction of glutinous rice moisture content with NIRS uses a self-developed FT-IR IPTEK T-1516 tool and the reference method used is the gravimetric method based on the Association of Official Analytical Chemists (AOAC). Data processing using Unsclambers software X version 10.5. The results showed that NIRS with the PCR method was able to produce a good model for estimating glutinous rice. This study produced four models of estimation of glutinous rice water content where one model was classified as very good performance (RPD 3) and three models were classified as good model performance (RPD 2) so that it could be said that all the models produced were suitable and good for estimating rice water content sticky rice. The best pretreatment in this study is the Standard Normal Variate (SNV) with an RPD value of 3.12, r of 0.95, R2 of 0.89, and RMSEC of 2.34. 


2019 ◽  
Vol 4 (3) ◽  
pp. 75-84
Author(s):  
Muslem Muslem ◽  
Sri Purnama Sari ◽  
Agus Arip Munawar

Abstrak, Parameter yang digunakan dalam penilaian mutu buah mangga antara lain ukuran atau berat, kekerasan, tingkat ketuaan serta bebas dari cacat. Kekerasan pada buah mangga merupakan fungsi dari tingkat kematangan, sedangkan kematangan berhubungan dengan tingkat ketuaan yang dapat diduga melalui penampilan visual. Vitamin C merupakan vitamin yang larut dalam air dan esensial untuk biosintesis kolagen.pengukuran vitamin C pada buah mangga menggunkan metode tetrasi, dan penggunaan gelombang elektromaknetik seperti Near Infrared. Penelitian ini bertujuan untuk memprediksi kadar vitamin C dalam buah mangga menggunakan metode Spektrofotometri UV-Vis dan Iodimetri, serta membandingkan hasil dari kedua metode tersebut. Sampel yang diidentifikasi yaitu buah mangga yang sudah matang dengan menggunakan model transformasi Attenuated Total Reflectance dan menggunakan metode Principal Component Analysis (PCA) dan menggunakan metode Principal Component Regression  (PCR). Penelitian ini menggunakan buah mangga jenis Arumanis, yang berjumlah 30 sampel. Prediksi vitamin C dengan NIRS menggunakan alat FT-IR IPTEK T-1516. Pengolahan data menggunakan Unscramble software® X versi 10.5. Hasil penelitian menunjukkan prediksi vitamin C mangga dengan metode Principal Component Regression (PCR) menghasilkan sufficient performance dengan nilai RPD yang didapat yaitu 2,0083 (r) sebesar 0,8638 , (R2 ) sebesar 0,7463 dan (RMSEC) sebesar 5,1854 Transformation Of Attenuated Total Reflectance (ATR) Near Infrared for prediction of Vitamin C In Arumanis Mangoes (Mangifera Indica)Abstract. Parameters used in assessing the quality of mangoes are size or weight, hardness, age level and free from defects. Hardness in mangoes is a function of maturity level, while the maturity is related to the level of aging that can be predicted through visual appearance. Vitamin C is a water-soluble vitamin which is essential for collagen biosynthesis. The measurement of vitamin C in mangoes use tetration methods, and the using of electromagnetic waves such as Near Infrared. This study aims to predict vitamin C contains in mango fruit using the UV-Vis and Iodymetry Spectrophotometry method, and comparing the results of the two methods. The samples identified were mature mangoes using the attenuated total reflectance transformation model and using the Principal Component Analysis (PCA) method also using the Principal Component Regression (PCR) method. This study used Arumanis mangoes, which amounted to 30 samples. Prediction of vitamin C with NIRS using the FT-IR IPTEK T-1516. Data processing use the Unscramble software® X 10.5 version. The results showed that the prediction of vitamin C mango using the Principal Component Regression (PCR) method resulted in sufficient performance with the obtained RPD value of  2,0083, (r) of 0,8638, (R2) of 0,7463 and (RMSEC) of 5,1854.


2012 ◽  
Vol 599 ◽  
pp. 151-154 ◽  
Author(s):  
Xiao Li Li ◽  
Xiao Peng Li ◽  
Yu Li

The Quantitative Structure Activity Relationship (QSAR) was used to correlate eleven physical and chemical properties (GTCi, Vc, BP, MP, Hf, Tc, Pc, MW, MV, logKOWand logKOC) with toxicity of polycyclic aromatic hydrocarbons (PAHs). A multi-parameter regression model was conducted to simulate the toxicity of PAHs after minimization of the multicollinearity among the ion characteristics using principal component regression (PCR). The toxicity of PAHs increased with the positively correlated variables including GTCi, Vc, BP, MW, MV, logKOWand logKOC. The regression model provided the high simulate ability, with Nash-Suttcliffe simulation efficiency coefficients (NSC) of 0.89 for the modeling. The model may be successfully employed to predict the toxicity of PAHs and be used for further analysis.


2013 ◽  
Vol 341-342 ◽  
pp. 797-800
Author(s):  
Xin Ma ◽  
Rong Guang Sun ◽  
Yong Feng Dong

The paper presents the design of vision system of the mobile robot, and shows methods of object recognition based on color images in the system of mobile robot. To adapt to the different light conditions, HSI color space is used. The system is able to meet the demand on rapidity and veracity of system of mobile robot. Experiment results show that the proposed technique is liable to accomplish object recognition in presence of changing illumination environment conditions.


2018 ◽  
Vol 124 ◽  
pp. 180-196 ◽  
Author(s):  
Shuichi Kawano ◽  
Hironori Fujisawa ◽  
Toyoyuki Takada ◽  
Toshihiko Shiroishi

Sign in / Sign up

Export Citation Format

Share Document