Article Per Year (5 Year)
p-Index From 2019 - 2024
0.444
P-Index
This Author published in this journals
All Journal Bulletin of Computer Science Research Journal on Education and Teacher Professionalism
Melly Br Bangun
Unknown Affiliation
Religion Computer Science & IT Education

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Bulletin of Computer Science Research

 1 
Penerapan Probabilistic Neural Network pada Klasifikasi Patogen Daun Bibit Jabon Berdasarkan Ciri Morfologi Spora Melly Br Bangun; Yeni Herdiyeni; Elis Nina Herliyana; Rossy Nurhasanah
Bulletin of Computer Science Research Vol. 4 No. 2 (2024): Februari 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bulletincsr.v4i2.325

Abstract

The aim of this research is to clasify pathogen of Jabon’s leaf seedling based on spora morphological features using Probabilistic Neural Network classifier. Three types of pathogen to be classified are Colletotrichum sp., Curvularia sp., and Fusarium sp.. The methodologies used are data acquisition using optilab camera microscope to obtain microscopic image data , preprocessing (grayscale, median smoothing, thresholding Otsu, region filling, median smoothing and dilate), morphology feature extraction (area, perimeter, area convex, convex perimeter, compactness, solidity, convexity and roundness), Probabilistic Neural Network classification, and evaluation. The basic morphological characteristics consisting of area, perimeter, convex area, convex perimeter, and derived morphological characteristics consisting of compactness, solidity, convexity and roundness. The experimental results of the morphological feature extraction showed that the compactness and roundness characteristics can be used to identify the three types of pathogens because with these characteristics each class of pathogen is separate. Testing for this research was carried out using 150 test data from three classes of objects from the dataset, namely class 1 (Colletotrichum sp.), class 2 (Curvularia sp.), and class 3 (Fusarium sp.). Then the results of pathogen classification using the application of the PNN algorithm in testing this research obtained an average accuracy value of 86.8% with a proportion of training data and test data of 80:20. The results of the PNN classification on 150 test data were that there were 36 data classified into Colletotrichum sp., 44 data classified into Curvularia sp., and 50 data classified into Fusarium sp. Further research could be done with the identification of digital microscopic images without cropping and systems that could clasify a colony image of pathogens clearly.
Co-Authors Elis Nina Herliyana Marisa Nabila Muhammad Adib Ahsani Oriza Salsabila Reny Furnawati Sitanggang Romi Anggun Zefanya Rossy Nurhasanah SANI SUSANTI