Claim Missing Document
Check
Articles

Found 4 Documents
Search

PENERAPAN MULTILAYER PERCEPTRON UNTUK KLASIFIKASI JENIS KULIT SAPI TERSAMAK Nunik Purwaningsih
Jurnal Teknoif Teknik Informatika Institut Teknologi Padang Vol 4 No 1 (2016): JURNAL TEKNOIF ITP
Publisher : ITP Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (337.681 KB) | DOI: 10.21063/jtif.2016.V4.1.1-7

Abstract

Penelitian ini bertujuan menerapkan metode multilayer perceptron (MLP) untuk mengklasifikasikan jenis kulit sapi tersamak berdasarkan ciri teksturnya. Terdapat empat jenis kulit sapi tersamak yang dijadikan sampel yaitu kulit samak nabati, kulit samak semi krom, kulit boks, dan kulit pull up. Data yang digunakan terdiri atas 24 citra kulit nabati, 16 buah citra kulit semi krom, 12 citra kulit boks, dan 8 citra kulit pull up. Tingkat ketepatan klasifikasi mencapai 87,83%. Jenis kulit yang bisa diidentifikasi paling tepat adalah kulit pull up dengan tingkat akurasi 98,75%.
Klasifikasikan Jenis Cacat Kulit Menggunakan SMOTE-GoogLeNet Prananda, Alifia Revan; Frannita, Eka Legya; Pramitaningrum, Erlita; Hidayat, Anwar; Setiawan , Wawan Budi; Purwaningsih , Nunik
JITU Vol 8 No 1 (2024)
Publisher : Universitas Boyolali

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Deep learning has been proven to be able to provide significant contributions to several fields, including industry. It has also been proven that it has resulted in an outstanding performance for classification, detection, and even segmentation processes. In the leather industry, it also successfully gave valuable results, especially for the leather defect inspection process. However, despite its outstanding performance, it remained a drawback because it produced insignificant results if employed in a small or imbalanced dataset. This research work focuses on the analysis of the implementation of the data balancing method for improving the performance of the deep learning method for classifying the types of leather defects. This research work was done by employing three processes. In the first step, we utilized the data balancing method to balance the data proportion. In the next step, we employed GoogLeNet as a deep learning architecture for training and testing processes. Our experiment was conducted in two scenarios. The first scenario was done by using the original dataset. Whereas the second scenario was accomplished by utilizing the data balancing method before training and testing. According to the experiment results, implementing the data balancing method successfully increased the performance of the deep learning method by more than 15%. It can be inferred that the proportion or the number of data strongly affected the performance of deep learning models.
DEVELOPMENT OF A MULTIPLE MARKING SYSTEM ON SHOELAST: AN EFFICIENT AND ADAPTIVE SOLUTION FOR MULTI-DESIGN PRODUCTION Umi Ramadhanti; Nunik Purwaningsih
Berkala Penelitian Teknologi Kulit, Sepatu, dan Produk Kulit Vol. 24 No. 2 (2025): Berkala Penelitian Kulit, Sepatu, dan Produk Kulit
Publisher : Politek ATK Yogyakarta Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58533/1fzb6c84

Abstract

The lasting process in the footwear industry is essential for ensuring the accuracy and symmetry of upper placement on the shoelast, which directly affects the final shape and quality of the shoe. In many footwear industries, shoelasts are used without placement guides, leading operators to rely on visual estimation. This condition reduces accuracy, increases defects, and lowers efficiency, particularly when a single shoelast is used for multiple designs. This study aims to develop an adaptive and easily implemented multiple marking system on shoelasts to support various women’s footwear designs. The research combines needs analysis through observation of the lasting process with experimental testing of marking creation and application. Markings were designed based on upper dimensions, distinguished by color for each design, and applied using durable yet non-permanent acrylic markers. Testing on two shoe articles (A and B) evaluated upper-position accuracy, lasting time, and operator responses. Results indicate that the multiple marking system significantly improves accuracy in upper placement and reduces lasting time from 8–10 minutes to 5–7 minutes. Operators reported clearer guidance and easier adaptation. Limited production trials further showed defect reduction for article A from 18% to 10% and for article B from 28% to 8%. Overall, the study demonstrates that an adaptive multiple marking system on the shoelast is an effective and economical method to enhance accuracy, improve lasting performance, reduce defects, and maintain flexibility for multi-design footwear production.
Deteksi Defect Sepatu Berbasis Fine-Tuned CNN Menggunakan Arsitektur DenseNet untuk Mendukung Proses Quality Control Jamila Jamila; Eka Legya Frannita; Gilang Fatikhul Burhan; Nunik Purwaningsih; Windra Bangun Nuswantoro; Anwar Hidayat; Erlita Pramitaningrum; Totok Yulaidin; Alifia Revan Prananda
TIN: Terapan Informatika Nusantara Vol 7 No 3 (2026): August 2026
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/tin.v7i3.9712

Abstract

Indonesia is a leading global exporter of footwear. Despite successfully exports million pair of shoes, shoes industry in Indonesia faces significant challenges due to conventional quality control (QC) limitations. This study proposes a deep learning-based defect detection model to enhance the efficiency and accuracy of the QC process. This study uses primary datasets collected from Laboratory of PIDI 4.0 Politeknik ATK Yogyakarta and employed in four experimental scenarios. The dataset consists of 2,828 images comprising of two classes (good class and reject class). This study utilizes fine-tuned CNN using DenseNet architecture for distinguishing two classes. Before employing the fine-tuned CNN method, we separate the data into three components, which are training, validation and testing data, with proportion of 80:10:10. The results demonstrate that the integration of data augmentation and the Adam optimizer yielded the highest performance, achieving accuracy, precision, recall and F1-score of 0.9546, 0.9551, 0.9546, 0.9545, respectively. These findings suggest that implementing automated deep learning models has potential to reduce rejection rates by modernizing traditional inspection methods, which can strengthen the competitiveness of Indonesia’s footwear industry in the global market.