DETEKSI PELAT NOMOR BERBASIS ALGORITMA LSTM PADA OPTICAL CHARACTER RECOGNITION

WARDHANI, ELISA (2023) DETEKSI PELAT NOMOR BERBASIS ALGORITMA LSTM PADA OPTICAL CHARACTER RECOGNITION. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (COVER) 01 COVER.pdf Download (351kB) | Preview
Preview	Text (ABSTRAK) 02 ABSTRAK.pdf Download (20kB) | Preview
	Text (BAB I) 03 BAB 1.pdf Restricted to Registered users only Download (71kB)
	Text (BAB II) 04 BAB 2.pdf Restricted to Registered users only Download (99kB)
	Text (BAB III) 05 BAB 3.pdf Restricted to Registered users only Download (79kB)
	Text (BAB IV) 06 BAB 4.pdf Restricted to Registered users only Download (1MB)
	Text (BAB V) 07 BAB 5.pdf Restricted to Registered users only Download (21kB)
	Text (DAFTAR PUSTAKA) 08 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (171kB)
	Text (LAMPIRAN) 09 LAMPIRAN.pdf Restricted to Registered users only Download (1MB)

Abstract

Illegal Parking which has become a frequent occurrence in the Jakarta area, encourages system transformation which includes the addition of automatic parking system facilities based on the building infrastructure and facilities regulation. The development of an automatic parking system can utilize license plate number detection to minimize the need to manually input license plate numbers into the parking system. In this study, LSTM algorithm training is done and implemented on Optical Character Recognition to detect license plate numbers accurately. Based on the evaluation results, the LSTM algorithm has a good performance in detecting license plate numbers with an accuracy rate of 86,36%. However, the LSTM algorithm performance improved when implemented on Optical Character Recognition with an accuracy rate of 95,8%. Hence, based on the evaluation, the LSTM algorithm that has been implemented on Optical Character Recognition is considered a preferable choice in license plate number detection as it has a higher level of accuracy compared to the use of the LSTM algorithm alone. Keywords: License Plate Number Detection, LSTM, Optical Character Recognition Parkir liar yang semakin marak terjadi di wilayah Jakarta mendorong untuk dilakukannya pembenahan sistem dan penambahan fasilitas parkir dimana hal ini meliputi kelengkapan prasarana dan sarana bangunan yang termasuk sistem parkir otomatis. Pengembangan sistem parkir otomatis dapat dimulai dari memanfaatkan sistem deteksi pelat nomor sehingga tidak diperlukan pemasukan pelat nomor secara manual oleh petugas pada sistem parkir. Dalam penelitian ini dilakukan pelatihan algoritma LSTM yang selanjutnya diimplementasikan terhadap Optical Character Recognition untuk dapat mendeteksi pelat nomor secara akurat. Berdasarkan hasil evaluasi algoritma LSTM pada Optical Character Recognition, algoritma LSTM memiliki performa yang baik dalam mendeteksi pelat nomor yaitu dengan tingkat akurasi 86,36%, namun performa algoritma LSTM jauh lebih baik saat diimplementasikan terhadap Optical Character Recognition dimana tingkat akurasi meningkat menjadi 95,8%. Dengan demikian berdasarkan evaluasi dari penelitian ini, algoritma LSTM yang diimplementasikan pada Optical Character Recognition dianggap sebagai pilihan yang lebih baik untuk dapat mendeteksi pelat nomor karena memiliki tingkat akurasi yang lebih tinggi dibanding penggunaan algoritma LSTM secara eksklusif. Kata Kunci: Deteksi Pelat Nomor, LSTM, Optical Character Recognition

Actions (login required)

View Item