Identifikasi Citra Tanaman Obat Jenis Rimpang dengan Euclidean Distance Berdasarkan Ciri Bentuk dan Tekstur

Abstract

In the midst of the Covid-19 pandemic, increasing the body's immunity is very important. Some experts suggest consuming medicinal plants or herbs to boost immunity. In addition to being used as a cooking spice, this rhizome type plant turns out to have properties and benefits for health, especially to increase immunity. However, many people do not know and it is difficult to distinguish the type of rhizome plant. This type of rhizome plant can be identified based on the characteristics seen from the shape and texture. However, most people judge the type of rhizome has a shape that is difficult to distinguish. This study aims to determine the type of medicinal plant rhizome with Euclidean distance and extraction of shape and texture. Extraction of shape features using metric and eccentricity parameters. This parameter is considered to be able to recognize shape objects and can distinguish them from other objects. Meanwhile, texture feature extraction uses Gray Level Co-occurence Matrix (GLCM) with contrast, correlation, energy, and homogeneity parameters. For the identification process, Euclidean distance is used which serves to represent the level of two images that consider the distance value from Euclidean. From the results of the evaluation using a confusion matrix by calculating precision, recall and accuracy, it gets a precision value of 83%, recal 87% and an accuracy of 85%. These results indicate that the Euclidean distance and extraction of shape and texture features can identify the object image of medicinal plants with rhizome types well

Classification of Mint Leaf Types Based on the Image Using Euclidean Distance and K-Means Clustering with Shape and Texture Feature Extraction

“…K-Means is an approach to group n objects that are affected by the attributes entered in k partitions, where k is less than n [7]. The steps in the K-means Clustering algorithm begin by counting the number of groupings, and continue with counting the number of centroids.…”

“…In this study, the verification of the success rate of the method used reached 84.00%. Subsequent research on the classification of rhizome plant species using euclidean distance [7]. In this study, the performance of the model was tested using a confusion matrix with the results of 83% precision, 87% recal and 85% accuracy.…”

“…There are many implementations of digital image processing, one of which is used for image classification. Image classification can be defined as the process of grouping image pixels into classes of the same type, so that these classes can describe certain characteristics that can be distinguished from other classes [7]. One of the parameters that can be used for image classification is the Euclidean distance.…”

Klasifikasi Citra Daun Herbal Dengan Menggunakan Backpropagation Neural Networks Berdasarkan Ekstraksi Ciri Bentuk

“…Pada Gambar 6, menunjukkan bahwa rata-rata akurasi hasil uji akurasi yang menggambarkan kedekatan hasil pengujian atau rata-rata hasil uji dengan nilai sebenarnya diperoleh nilai sebesar 88,75%. Hasil tersebut kemudian dikonversi dalam kategori hasil klasifikasi dengan acuan sebagai berikut: Baik, dengan rentang nilai antara 76% hingga 100%; Cukup, dengan rentang nilai 56% hingga 75%; Kurang Baik, dengan rentang nilai 40% hingga 55%, sedangkan Kurang Baik, memiliki nilai kurang dari 40% [17]. Maka, akurasi dari model klasifikasi daun herbal menggunakan algoritma Backpropagation Neural Network (BNN) dengan ekstraksi ciri bentuk melalui parameter metric dan eccentricity masuk dalam kategori baik.…”

Identifikasi Kualitas Kesegaran Ikan Menggunakan Algoritma K-Nearest Neighbor Berdasarkan Ekstraksi Ciri Warna Hue, Saturation, dan Value (HSV)

“…Penelitian ini bertujuan untuk mengembangkan metode penelitian identifikasi kesegaran terhadap ikan nila menggunakan teknik citra digital, dimana penentuan kualitas kesegaran ikan berdasarkan nilai rata -rata grayscale, sedangkan dalam penelitian ini menggunakan nilai ekstraksi ciri warna Hue, Saturation, dan Value. Adapun metode klasifikasi yang digunakan dalam penelitian ini yaitu metode K-Nearest Neighbor dengan rumus jarak yang digunakan yaitu Eucledian Distance, dimana merepresentasikan tingkat kedekatan atau kemiripan antara jarak citra yang diuji dengan citra yang dilatih [10]. Parameter Hue, Saturation, dan Value berhasil diterapkan dalam menentukan kualitas kesegaran ikan dengan perolehan nilai akurasi untuk pelatihan sebesar 93% dan pengujian sebesar 90%.…”