Traditional Machine Learning-Based Classification of Cashew Kernels Using Colour Features

Geofrey Prudence Baitu; Omsalma Alsadig Adam Gadalla; Y. Benal Öztekin

Article Details

Traditional Machine Learning-Based Classification of Cashew Kernels Using Colour Features

Authors:

Geofrey Prudence Baitu | Omsalma Alsadig Adam Gadalla | Y. Benal Öztekin

Publisher:

Tekirdag Namik Kemal University

Journal:

Tekirdağ Ziraat Fakültesi Dergisi Vol.20 Issue 1, pp.115-124

Document Type:

Article

Publication Date:

1/2023

Document Language:

English

Document ID (Org No.):

2001281946

pISSN:

1302-7050

Keywords:

Abstract:

 Cashew is one of the major commercial commodities contributing to the national economy of Tanzania as foreign revenue. And yet still the processing of cashew is run locally using manual labour for a big part. If processed well under ideal conditions, cashews kernels are expected to be white in colour. But due to various factors like prolonged roasting in the steam chambers or over-drying, some cashew kernels tend to have a slight brown colour, and these are referred to as scorched cashews. Despite sharing the same characteristics with white cashew kernels, including nutritional quality, these cashew kernels are supposed to be graded differently. In many places around the world, particularly in Tanzania, the sorting and grading process of cashew kernels is performed by hand. In international trade, cashew grading is very important and this means more effective and consistent methods need to be applied in this stage of production in order to increase the quality of the products. The objective of this study was to evaluate the use of traditional Machine Learning techniques in the classification of cashew kernels as white or scorched by using colour features. In this experiment, various colour features were extracted from the images. The extracted features include the means μ, standard deviations σ, and skewness γ of the channels in RGB and HSV colour spaces. The relevant features for this classification problem were selected by applying the wrapper approach using the Boruta Library in Python, and the irrelevant ones were removed. 5 models are studied and their efficiencies analysed. The studied models are Logistic Regression, Decision Tree, Random Forest, Support Vector Machine and K-Nearest Neighbour. The Decision Tree model recorded the least accuracy of 98.4. The maximum accuracy of 99.8 was obtained in the Random Forest model with 100 trees. Due to simplicity in application and high accuracy, the Random Forest is recommended as the best model from this study.

 Cashew is one of the major commercial commodities contributing to the national economy of Tanzania as foreign revenue. And yet still the processing of cashew is run locally using manual labour for a big part. If processed well under ideal conditions, cashews kernels are expected to be white in colour. But due to various factors like prolonged roasting in the steam chambers or over-drying, some cashew kernels tend to have a slight brown colour, and these are referred to as scorched cashews. Despite sharing the same characteristics with white cashew kernels, including nutritional quality, these cashew kernels are supposed to be graded differently. In many places around the world, particularly in Tanzania, the sorting and grading process of cashew kernels is performed by hand. In international trade, cashew grading is very important and this means more effective and consistent methods need to be applied in this stage of production in order to increase the quality of the products. The objective of this study was to evaluate the use of traditional Machine Learning techniques in the classification of cashew kernels as white or scorched by using colour features. In this experiment, various colour features were extracted from the images. The extracted features include the means μ, standard deviations σ, and skewness γ of the channels in RGB and HSV colour spaces. The relevant features for this classification problem were selected by applying the wrapper approach using the Boruta Library in Python, and the irrelevant ones were removed. 5 models are studied and their efficiencies analysed. The studied models are Logistic Regression, Decision Tree, Random Forest, Support Vector Machine and K-Nearest Neighbour. The Decision Tree model recorded the least accuracy of 98.4. The maximum accuracy of 99.8 was obtained in the Random Forest model with 100 trees. Due to simplicity in application and high accuracy, the Random Forest is recommended as the best model from this study.

 Cashew is one of the major commercial commodities contributing to the national economy of Tanzania as foreign revenue. And yet still the processing of cashew is run locally using manual labour for a big part. If processed well under ideal conditions, cashews kernels are expected to be white in colour. But due to various factors like prolonged roasting in the steam chambers or over-drying, some cashew kernels tend to have a slight brown colour, and these are referred to as scorched cashews. Despite sharing the same characteristics with white cashew kernels, including nutritional quality, these cashew kernels are supposed to be graded differently. In many places around the world, particularly in Tanzania, the sorting and grading process of cashew kernels is performed by hand. In international trade, cashew grading is very important and this means more effective and consistent methods need to be applied in this stage of production in order to increase the quality of the products. The objective of this study was to evaluate the use of traditional Machine Learning techniques in the classification of cashew kernels as white or scorched by using colour features. In this experiment, various colour features were extracted from the images. The extracted features include the means μ, standard deviations σ, and skewness γ of the channels in RGB and HSV colour spaces. The relevant features for this classification problem were selected by applying the wrapper approach using the Boruta Library in Python, and the irrelevant ones were removed. 5 models are studied and their efficiencies analysed. The studied models are Logistic Regression, Decision Tree, Random Forest, Support Vector Machine and K-Nearest Neighbour. The Decision Tree model recorded the least accuracy of 98.4. The maximum accuracy of 99.8 was obtained in the Random Forest model with 100 trees. Due to simplicity in application and high accuracy, the Random Forest is recommended as the best model from this study.

 Kaju, Tanzanyanın ülke ekonomisine dış gelir olarak katkı sağlayan başlıca ticari ürünlerden biridir. Kaju çekirdeklerinin işlenmesi, halen büyük ölçüde el emeği kullanılarak yerel olanaklarla yapılmaktadır. İdeal koşullarda iyi işlenirse kajuların beyaz renkte olması beklenir. Ancak, buhar odalarında uzun süre kavurma veya aşırı kurutma gibi çeşitli faktörler nedeniyle, bazı kaju çekirdekleri hafif kahverengi bir renge dönüşebilmektedir. Renk değiştirmiş bu kajulara kavrulmuş kaju denir. Besin kalitesi de dahil olmak üzere beyaz kaju çekirdekleri ile aynı özelliklere sahip olmasına rağmen, renk ve görünüm tüketicilerin kalite algısını etkilediği için bu kaju çekirdeklerinin ayrılması gerekmektedir. Tanzanya başta olmak üzere dünyanın pek çok yerinde kaju çekirdeklerinin ayırma ve sınıflandırma işlemi elle yapılmaktadır. Uluslararası ticarette, kaju sınıflandırması çok önemli olup ürün kalitesini artırmak için üretimin bu aşamasında daha etkili ve tutarlı yöntemlerin uygulanması gerektiği anlamına gelir. Bu çalışmanın amacı, kaju çekirdeklerinin beyaz veya kavrulmuş olarak sınıflandırılmasında renk özellikleri kullanılarak geleneksel Makine Öğrenmesi tekniklerinin kullanımının değerlendirilmesidir. Bu çalışmada, görüntülerden farklı renk özellikleri çıkarılmıştır. Çıkarılan özellikler, RGB ve HSV renk uzaylarında kanalların ortalamaları μ, standart sapmaları σ ve çarpıklığını γ içerir. Pythonda Boruta Kütüphanesi kullanılarak sarmal wrapper yöntemi uygulanarak bu sınıflandırma problemi için ilgili özellikler seçilmiş ve ilgili olmayanlar çıkarılmıştır. Bu çalışmada 5 model çalışılmış ve verimlilikleri analiz edilmiştir. Değerlendirme teknikleri Lojistik Regresyon, Karar Ağacı, Rastgele Orman, Destek Vektör Makinesi ve K-En Yakın Komşu KNN yöntemleridir. Karar Ağacı modeli, 98,4 ile en düşük doğruluğu vermiştir. 100 ağaçlı Rastgele Orman modelinde maksimum 99,8 doğruluk elde edilmiştir. Uygulamadaki basitliği ve yüksek doğruluğu nedeniyle Rastgele Orman bu çalışma için en iyi model olarak önerilmektedir.

Article Details

Traditional Machine Learning-Based Classification of Cashew Kernels Using Colour Features

Authors:

Geofrey Prudence Baitu | Omsalma Alsadig Adam Gadalla | Y. Benal Öztekin

Publisher:

Tekirdag Namik Kemal University

Journal:

Tekirdağ Ziraat Fakültesi Dergisi Vol.20 Issue 1, pp.115-124

Document Type:

Article

Publication Date:

1/2023

Document Language:

English

Document ID (Org No.):

2001281946

pISSN:

1302-7050

Logistic Regression

Decision Tree

Random Forest

Cashews

Support Vector Machine K-Nearest Neighbour

Lojistik Regresyon

Karar Ağacı

Rastgele Orman

Destek Vektör Makinesi ve K-En Yakın Komşu (KNN)

Kaju fıstığı

Issues

Article Details

Authors: Geofrey Prudence Baitu | Omsalma Alsadig Adam Gadalla | Y. Benal Öztekin

Publisher: Tekirdag Namik Kemal University

Journal: Tekirdağ Ziraat Fakültesi Dergisi Vol.20 Issue 1, pp.115-124

Document Type: Article

Publication Date: 1/2023

Document Language: English

Document ID (Org No.): 2001281946

pISSN: 1302-7050

Issues

Authors:

Geofrey Prudence Baitu | Omsalma Alsadig Adam Gadalla | Y. Benal Öztekin

Publisher:

Tekirdag Namik Kemal University

Journal:

Tekirdağ Ziraat Fakültesi Dergisi Vol.20 Issue 1, pp.115-124

Document Type:

Article

Publication Date:

1/2023

Document Language:

English

Document ID (Org No.):

2001281946

pISSN:

1302-7050