Article Per Year (5 Year)
p-Index From 2020 - 2025
0.408
P-Index
This Author published in this journals
All Journal Teknosains : Jurnal Sains,Teknologi dan Informatika
Hermawan Syahputra
Medan State University, Medan, North Sumatra, Indonesia 20221
Agriculture, Biological Sciences & Forestry Automotive Engineering Computer Science & IT

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

Found 2 Documents
Search

 1 
Red Chili Classification Using HSV Feature Extraction and Naive Bayes Classifier Hermawan Syahputra; Josua Nainggolan; Johanes Apriadi Parlinggoman Sirait; Muhammad Fadlan Ikromi; Putri Ameliya Lubis
TEKNOSAINS : Jurnal Sains, Teknologi dan Informatika Vol 11 No 1 (2024): TEKNOSAINS: Jurnal Sains, Teknologi dan Informatika
Publisher : LPPMPK-Sekolah Tinggi Teknologi Muhammadiyah Cileungsi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37373/tekno.v11i1.593

Abstract

In the culinary industry, the classification of red chili pepper types is used to identify varieties that differ in terms of flavor, pungency, or other uniqueness. This enables their proper use in various recipes and meals. In the market, the classification of red chili pepper types helps in pricing, variety selection, or quality standards applied. For this reason, the purpose of this research is to classify red chili peppers using HSV Feature Extraction and Naive Bayes Clasifier. The stages carried out include: data collection, preprocessing, feature extraction and classification. Red chilies are grouped into 4 classes, namely large red chilies, cakplak red chilies, curly red chilies and chili red chilies. The red chili data used is 119 training data and 123 testing data. In the preprocessing, the image is converted to grayscale, then converted to binary image with the thresholding method. Furthermore, feature extraction is done with the HSV method. Finally, classification is done with Naive Bayes. The results of the study provide an accuracy value for training data of 92.43% and for testing data obtained an accuracy of 92.69%. This method is suitable for use in classification because it gives good results
Detection of mango leaf disease using the convolution neural network method Vinny Ramayani Saragih; Nur Azizi; Alfattah Atalarais; Reza Ananda Hatmi; Hermawan Syahputra
TEKNOSAINS : Jurnal Sains, Teknologi dan Informatika Vol 11 No 1 (2024): TEKNOSAINS: Jurnal Sains, Teknologi dan Informatika
Publisher : LPPMPK-Sekolah Tinggi Teknologi Muhammadiyah Cileungsi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37373/tekno.v11i1.639

Abstract

Mango trees are a common plant in Indonesia. In 2021, 2.84 million tonnes of mangoes were produced in Indonesia, according to the Central Statistics Agency (BPS). You can eat mangoes either ripe or unripe. Additionally, this fruit can be turned into meals and beverages. Many farmers grow mango plants, however, Some illnesses can infect mangoes and lead to crop failure or poor fruit quality. Fungal infections like anthracnose and black fungus, often known as black fungus, affect mango plants, but many farmers continue to mistakenly believe they are identifying plant illnesses and pests. To the type of disease in mango plants, the Convolutional Neural Network (CNN) method was applied in this study. It has been demonstrated that CNN is a very efficient way of processing images and identifying key elements for intricate pattern recognition. A total of 1,405 leaf photos from three different categories—525 anthracnose images, 656 black sooty mold images, and 224 healthy images—were used as samples for CNN to identify illnesses in mango plants. This image data was taken from the kaggle.com website. The CNN model is trained using distinct datasets into training data and validation data after data collection and preprocessing. On training data, the model is 95% accurate, while on validation data, it is 98% accurate. By feeding photos of mango leaves into the model and evaluating the predictions, the detection is put into practice. Action can be taken to control the illness in these mango trees based on prediction findings showing the presence of disease with a decent amount of confidence
Co-Authors Alfattah Atalarais Johanes Apriadi Parlinggoman Sirait Josua Nainggolan Muhammad Fadlan Ikromi Nur Azizi Putri Ameliya Lubis Reza Ananda Hatmi Vinny Ramayani Saragih