Segmentasi Citra Formulir Menggunakan Bounding box untuk Pengambilan Objek Gambar
DOI:
https://doi.org/10.52436/1.jpti.1071Kata Kunci:
aksara OKU timur, bounding box, cropping, pengenalan karakter, segmentasi citra, warisan budayaAbstrak
Penelitian ini bertujuan mengembangkan metode segmentasi citra berbasis bounding box dan teknik cropping untuk mengisolasi karakter aksara OKU Timur sebagai upaya pelestarian budaya lokal. Data dikumpulkan melalui kuesioner yang melibatkan 102 responden, masing-masing menulis karakter aksara pada lembar khusus. Eksperimen dilakukan pada lima sampel gambar yang berisi karakter serupa, namun ditulis oleh individu berbeda, guna menguji konsistensi dan ketahanan metode terhadap variasi tulisan tangan. Proses segmentasi dievaluasi menggunakan metrik kuantitatif, yaitu precision, recall, F1-score dan akurasi, dengan hasil rata-rata precision 71,76%, recall 78,33%, F1-score 74,9%, dan akurasi 78,33%. Hasil terbaik mencapai akurasi 100%, sedangkan hasil terendah 33,33%, menunjukkan adanya variasi tingkat keberhasilan segmentasi. Temuan ini menegaskan bahwa pendekatan yang diusulkan cukup efektif dalam mengidentifikasi karakter aksara meskipun terdapat perbedaan gaya penulisan. Kontribusi utama penelitian ini adalah menyediakan solusi digitalisasi aksara tradisional berbasis pengolahan citra, yang dapat mendukung upaya pelestarian dan pengembangan teknologi pengenalan karakter untuk aksara daerah.
Unduhan
Referensi
Y. Sewell, “Linguistic pragmatism, lingua francae, and language death in Indonesia,” J. Lang. Teach., vol. 2, no. 11, pp. 15–19, 2022, doi: 10.54475/jlt.2022.015.
R. F. Kusumaningtyas, A. Hidayat, G. P. Soebiakto, A. F. Permana, and I. H. Abdullah, “Traditional Cultural Expression as an Embodiment of Indigenous Communities and Regional Identity (Semarang Indonesia Case),” J. Indones. Leg. Stud., vol. 8, no. 1, pp. 45–92, 2023, doi: 10.15294/jils.v8i1.63191.
P. Mikaresti and H. Mansyur, “Pewarisan Budaya Melalui Tari Kreasi Nusantara,” Gorga J. Seni Rupa, vol. 11, no. 1, p. 147, 2022, doi: 10.24114/gr.v11i1.33333.
S. Ivanov, “Modern Technologies in the Study, Preservation and Management of Cultural Heritage,” J. Sci. Appl. Res., vol. 23, no. 1, pp. 5–25, 2022, doi: 10.46687/jsar.v23i1.349.
A. Dhar, “Rich Cultural Traditions to Be Preserved Through Digitization- A Task of Great Responsibility,” Turkish Online J. Qual. Inq., vol. 11, no. 1, pp. 415–422, 2023, doi: 10.52783/tojqi.v11i1.9975.
A. R. Himamunanto, “Restorasi Digital Pada Model Kerusakan Citra Aksara Jawa Cetak,” J. Teknol. Informasi-Aiti |, pp. 193–199, 2016.
P. Rosyani, R. Amalia, and I. H. Ikasari, “Deteksi Objek dengan Model Warna Ycbcr dan Similiarity Distance,” J. Sist. dan Teknol. Inf., vol. 9, no. 2, p. 98, 2021, doi: 10.26418/justin.v9i2.44230.
A. Syahfaridzah, A. K. Panggabean, and N. A. Ardiningsih, “Mendeteksi Secara Otomatis Objek Gerakan Berdasarkan Gaussian Mixture Model Menggunakan Aplikasi Matlab,” Method. J. Tek. Inform. dan Sist. Inf., vol. 6, no. 2, pp. 19–23, 2020, doi: 10.46880/mtk.v6i2.242.
J. Pravalika, S. Korla, E. Kavya, and Sravani, “Handwritten Character Recognition to Obtain Editable Text,” E3S Web Conf., vol. 391, pp. 1–6, 2023, doi: 10.1051/e3sconf/202339101059.
S. Patil et al., “Enhancing Optical Character Recognition on Images with Mixed Text Using Semantic Segmentation,” J. Sens. Actuator Networks, vol. 11, no. 4, pp. 1–20, 2022, doi: 10.3390/jsan11040063.
W. Li, W. Liu, J. Zhu, M. Cui, X. S. Hua, and L. Zhang, “Box-Supervised Instance Segmentation with Level Set Evolution,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 13689 LNCS, pp. 1–18, 2022, doi: 10.1007/978-3-031-19818-2_1.
O. V. Chávez, J. Flores–Troncoso, J. U. M. Minjares, R. O. Reyna, E. G. Sánchez, and R. O. Reyna, “Image segmentation of Capsicum annuum chili with lighting problems using the otsu method,” Stud. Eng. Exact Sci., vol. 3, no. 4, pp. 560–573, 2022, doi: 10.54021/sesv3n4-001.
P. Soille and P. Vogt, “Morphological Spatial Pattern Analysis: Open Source Release,” Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. - ISPRS Arch., vol. 48, no. 4/W1-2022, pp. 427–433, 2022, doi: 10.5194/isprs-archives-XLVIII-4-W1-2022-427-2022.
S. Koda and I. Morikawa, “Bounding-box Watermarking: Defense against Model Extraction Attacks on Object Detectors,” no. i, pp. 15–17, 2024, [Online]. Available: http://arxiv.org/abs/2411.13047
Y. Wang and Z. Liao, “A Method for Object Extraction from Crop Image Based on Visual Saliency,” J. Phys. Conf. Ser., vol. 2171, no. 1, 2022, doi: 10.1088/1742-6596/2171/1/012007.
Fathima Chandhini S, Rashad H, Gowseelan K, and Jayasarathy S, “Extraction of Character from Visuals and Images Using OpenCV,” Int. J. Sci. Res. Comput. Sci. Eng. Inf. Technol., vol. 3307, pp. 194–200, 2023, doi: 10.32628/cseit2390363.
M. Boillet, C. Kermorvant, and T. Paquet, “Robust text line detection in historical documents: learning and evaluation methods,” Int. J. Doc. Anal. Recognit., vol. 25, no. 2, pp. 95–114, 2022, doi: 10.1007/s10032-022-00395-7.
D. Peng, L. Jin, Y. Liu, C. Luo, and S. Lai, “PageNet: Towards End-to-End Weakly Supervised Page-Level Handwritten Chinese Text Recognition,” Int. J. Comput. Vis., vol. 130, no. 11, pp. 2623–2645, 2022, doi: 10.1007/s11263-022-01654-0.
T. Ahmed, M. Uddin, M. A. R. Khan, and A. R. M. Hasan, “Offline Handwritten Character Recognition Including Compound Character from Scanned Document,” Asian J. Res. Comput. Sci., vol. 14, no. 4, pp. 119–129, 2022, doi: 10.9734/ajrcos/2022/v14i4297.
S. Luo, X. Li, and X. Zhang, “Bounding-box deep calibration for high performance face detection,” IET Comput. Vis., vol. 16, no. 8, pp. 747–758, 2022, doi: 10.1049/cvi2.12122.
I. Menggunakan and A. Yulianto, “Pengembangan Sistem Pengenalan Plat Nomor Pendahuluan Tinjauan pustaka,” vol. 23, pp. 571–578, 2024.
L. Abdiansah, A. Eviyanti, and N. L. Azizah, “Implementation of Convolutional Neural Networks Algorithm for Javanese Handwriting Recognition Penerapan Algoritma Convolutional Neural Networks untuk Pengenalan Tulisan Tangan Aksara Jawa,” vol. 5, no. April, pp. 496–504, 2025.
