Article Per Year (5 Year)
p-Index From 2020 - 2025
0.61
P-Index
This Author published in this journals
All Journal JUITA : Jurnal Informatika Sistemasi: Jurnal Sistem Informasi Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
Helmawati, Nita
Unknown Affiliation
Computer Science & IT Electrical & Electronics Engineering Engineering

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Analysis for Detecting Banana Leaf Disease Using the CNN Method Helmawati, Nita; Utami, Ema
JUITA: Jurnal Informatika JUITA Vol. 13 Issue 1, March 2025
Publisher : Department of Informatics Engineering, Universitas Muhammadiyah Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30595/juita.v13i1.24514

Abstract

Banana farmers face major challenges due to banana leaf diseases such as Cordana, Pestalotiopsis and Sigatoka, which severely affect the quality and quantity of the crop. Early detection of these diseases is particularly challenging as the initial symptoms are often similar to other disorders. To solve this problem, fast and accurate automated detection is needed to help farmers effectively identify diseases on banana leaves. This research focuses on developing a banana leaf disease detection model using Convolutional Neural Network (CNN) method with MobileNetV2 architecture. The dataset used consists of 937 images of both infected and healthy banana leaves. These images were collected under various lighting conditions and viewing angles to simulate real field situations. The dataset was divided into 70% for training, 20% for validation, and 10% for testing, to ensure robust model evaluation. The CNN model was trained to recognize important visual features on banana leaves that indicate disease infection. The results showed that the model was able to detect banana leaf diseases with an accuracy of 90.62%, indicating high effectiveness. This accuracy confirms the potential of CNN in significantly improving the disease detection process on banana plants. This research is expected to help farmers identify diseases more quickly and accurately, thereby minimizing yield losses and increasing productivity. In addition, this research provides valuable insights into the application of technology in agriculture, particularly in plant disease detection which opens up opportunities for further advancements in this sector.
Optimizing the Profile Matching Algorithm using the Analytical Hierarchy Process in the Selection of Teaching Assistants Helmawati, Nita; Norhikmah, Norhikmah
Sistemasi: Jurnal Sistem Informasi Vol 12, No 3 (2023): Sistemasi: Jurnal Sistem Informasi
Publisher : Program Studi Sistem Informasi Fakultas Teknik dan Ilmu Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v12i3.3172

Abstract

The selection of the best practicum assistants is traditionally done through a conventional method, which involves voting by active students attending lab classes. However, upon evaluation, it was found that the results were not accurate. Some cases revealed that assistants were chosen based solely on popularity or recognition among the students, possibly influenced by physical appearance or public speaking skills in front of the class, while other important aspects were not considered. This situation could lead to social jealousy. The problem lies in the difficulty of combining evaluation criteria and determining the relative weights for each criterion in the process of selecting the best practicum assistants at the college. Additionally, there is a lack of objectivity in decision-making during the selection process, resulting in an unstructured and immature decision-making process. Therefore, this research aims to enhance the process of selecting the best practicum assistants at the college through optimizing the profile matching algorithm using the Analytic Hierarchy Process (AHP) method. AHP's role involves checking the weights and making paired comparisons to evaluate each criterion and determine the criterion weights. AHP is also utilized to ensure consistency in determining the weights. On the other hand, the role of profile matching is to provide accurate rankings or comparisons based on the suitability scores between the profiles of potential assistants and the reference profile. The combination of these two algorithms is expected to result in a more accurate selection of practicum assistants by effectively measuring the decision criteria weights. Therefore, the difficulty of combining evaluation criteria and determining the relative weights for each criterion can be minimized. Furthermore, optimizing the profile matching algorithm will enable a more objective decision-making process for selecting the best practicum assistants through more accurate rankings or comparisons based on the suitability scores with the reference profile. Based on this optimization, the collaboration of the two algorithms can achieve comparison results with an accuracy rate of 90%.
Deep Learning-Based Soybean Leaf Disease Classification Using DenseNet121, Xception, and MobileNetV2 Helmawati, Nita; Buana, Yopy Tri; Darmawan, Eko Rahmad; Kusrini, Kusrini
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 9 No 6 (2025): December 2025 (in progress)
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v9i6.6271

Abstract

This study is driven by the challenge of soybean leaf diseases, which significantly reduce agricultural productivity and pose a threat to food security. To address this issue, we developed a deep learning–based classification model for soybean leaf disease detection, employing three prominent architectures: DenseNet121, Xception, and MobileNetV2. The dataset comprised 770 images representing six disease categories and one healthy category, which was expanded to 5,880 images using data augmentation techniques. The dataset was evaluated under three experimental scenarios with splits of 70% training, 10% validation, and 20% testing. Experimental results demonstrated that the DenseNet121 model, optimized with AdamW, achieved the highest accuracy at 90.14%, outperforming MobileNetV2 (85.48%) and Xception (65.37%). Moreover, DenseNet121 exhibited the most consistent performance in classifying the diverse categories of soybean leaf diseases.
Co-Authors Buana, Yopy Tri Darmawan, Eko Rahmad Ema Utami Kusrini, Kusrini Norhikmah Norhikmah