Image classification involves categorizing an image's pixels into specific classes based on their unique characteristics. It has diverse applications in everyday life. One such application is the classification of diseases on corn leaves. Corn is a widely consumed staple food in Indonesia, and healthy corn plants are crucial for meeting market demands. Currently, disease identification in corn plants relies on manual checks, which are time-consuming and less effective. This research aims to automate disease identification on corn leaves using the Support Vector Machine (SVM), K-Nearest Neighbor (K-NN) with K=2, and Stochastic Gradient Descent (SGD) algorithms. The classification process utilizes the Histogram of Oriented Gradients (HOG) feature extraction method with a dataset of corn leaf images. The classification results achieved an accuracy of 71.44%, AUC of 79.16%, precision of 70.08%, recall of 71.44%, and f1 score of 67.11%. The highest accuracy was obtained by combining HOG feature extraction with the SGD algorithm.
<p class="Abstrak">Nilai yang hilang membutuhkan preprosesing dengan teknik imputasi untuk menghasilkan data yang lengkap. Proses imputasi membutuhkan initial bobot yang sesuai, karena data yang dihasilkan adalah data pengganti. Pemilihan nilai bobot yang optimal dan kesesuaian nilai <em>K</em> pada metode <em>K-Means</em> Imputation (KMI) merupakan masalah besar, sehingga menimbulkan error semakin meningkat. Model gabungan algoritma genetika (GA) dan KMI atau yang dikenal GAKMI digunakan untuk menentukan bobot optimal pada setiap <em>cluster</em> data yang mengandung nilai yang hilang. Algoritma genetika digunakan untuk memilih bobot dengan menggunakan pengkodean bilangan riel pada kromosom. Model hybrid GA dan KMI dengan pengelompokan menggunakan jumlah jarak <em>Euclidian</em> setiap titik data dari pusat clusternya. Pengukuran kinerja algoritma menggunakan fungsi kebugaran optimal dengan nilai MSE terkecil. Hasil percobaan data hepatitis menunjukkan bahwa GA efisien dalam menemukan nilai bobot awal optimal dari ruang pencarian yang besar. Hasil perhitungan menggunakan nilai MSE =0.044 pada K=3 dan replika ke-5 menunjukkan kinerja GAKMI menghasilkan tingkat kesalahan yang rendah untuk data hepatitis dengan atribut campuran. Hasil penelitian dengan menggunakan pengujian tingkat imputasi menunjukkan algoritma GAKMI menghasilkan nilai <em>r</em> = 0.526 lebih tinggi dibandingkan dengan metode lainnya. Penelitian ini menunjukkan GAKMI menghasilkan nilai r yang lebih tinggi dibandingkan metode imputasi lainnya sehingga dianggap paling baik dibandingkan teknik imputasi secara umum. </p><p class="Abstrak"> </p><p class="Abstrak"><em><strong>Abstract</strong></em></p><p class="Judul2"><em>Missing values require preprocessing techniques as imputation to produce complete data. Complete data imputation results require the appropriate initial weights, because the resulting data is replacement data. The choice of the optimal weighting value and the suitability of the network nodes in the K-Means Imputation (KMI) method are big problems, causing increasing errors. The combined model of Genetic Algorithm (GA) and KMI is used to determine the optimal weights for each data cluster containing missing values. Genetic algorithm is used to select weights by using real number coding on chromosomes. GA is applied to the KMI using clustering calculated using the sum of the Euclidean distances of each data point from the center of the cluster. Performance measurement algorithms using the fitness function optimally with the smallest MSE value. The results of the hepatitis data experiment show that GA is efficient in finding the optimal initial weight value from a large search space. The results of calculations using the MSE value = 0.04 </em><em>for</em><em> K = 3 and the 5th replication</em><em>. So, </em><em>GAKMI resulted in a low error rate for mixed data. The results of research using imputation level testing </em><em>performed</em><em> GAKMI produc</em><em>ed</em><em> r = 0.526 higher than the other methods. Thus, the higher the r value, the best for the imputation technique.</em></p><p class="Abstrak"><em><strong><br /></strong></em></p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.