KLASIFIKASI TANAMAN BIDARA BERDASARKAN TEKSTUR DAUN MENGGUNAKAN METODE GRAY LEVEL CO-OCCURANCE MATRIX (GLCM) DAN ALGORITMA SUPPORT VECTOR MACHINE (SVM)
Total View This Week0
Total View This Week0
Institusion
Universitas Pembangunan Nasional Veteran Jakarta
Author
Adrian Budi Prawira, .
Subject
QA75 Electronic computers. Computer science
Datestamp
2021-04-06 07:38:16
Abstract :
Bidara is a type of fruit-producing plant that grows in dry areas. Based on the species, there are four types of plants bidara scattered in several places, namely the Bidara Arab (Ziziphus spina- christi), Bidara Upas (Merremia mammosa Hall.f.), Chinese Bidara (Ziziphus mauritiana Lam.), and Sea Bidara (Strychnos lucida R.Br.). Many people are looking for this plant to be cultivated, for daily consumption, and traded for treatment. However, there are still many people who are not well informed about how to distinguish between these plant species. With this problem, a solution is needed in order to minimize the error rate in distinguishing the types of species in bidara plants. The use of image processing can help in observing the texture of the leaves of bidara. In this study, a classification model will be made which functions to distinguish the types of bidara plants using the Support Vector Machine algorithm. Meanwhile, the extraction of texture features on bidara leaves will be observed using the Gray Level Co- occurrence Matrix method. The results of the research that have been carried out have produced a performance that does not disappoint in the detection of bidara plants. With the use of the Polynomial Quadratic SVM kernel, the best average obtained results were 84% accuracy, 92% precision, and 79.67% recall.
Bidara is a type of fruit-producing plant that grows in dry areas. Based on the species, there are four types of plants bidara scattered in several places, namely the Bidara Arab (Ziziphus spina- christi), Bidara Upas (Merremia mammosa Hall.f.), Chinese Bidara (Ziziphus mauritiana Lam.), and Sea Bidara (Strychnos lucida R.Br.). Many people are looking for this plant to be cultivated, for daily consumption, and traded for treatment. However, there are still many people who are not well informed about how to distinguish between these plant species. With this problem, a solution is needed in order to minimize the error rate in distinguishing the types of species in bidara plants. The use of image processing can help in observing the texture of the leaves of bidara. In this study, a classification model will be made which functions to distinguish the types of bidara plants using the Support Vector Machine algorithm. Meanwhile, the extraction of texture features on bidara leaves will be observed using the Gray Level Co- occurrence Matrix method. The results of the research that have been carried out have produced a performance that does not disappoint in the detection of bidara plants. With the use of the Polynomial Quadratic SVM kernel, the best average obtained results were 84% accuracy, 92% precision, and 79.67% recall.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
