INDONESIAN LICENSE PLATE RECOGNITION WITH IMPROVED HORIZONTAL-VERTICAL EDGE PROJECTION

NUDUDDIN, IMAM (2020) INDONESIAN LICENSE PLATE RECOGNITION WITH IMPROVED HORIZONTAL-VERTICAL EDGE PROJECTION. S1 thesis, Universitas Mercu Buana Jakarta.

Text (JURNAL MAHASISWA) Jurnal_41516010049_Imam Nududdin.pdf Restricted to Registered users only Download (2MB)

Abstract

License Plate Recognition (LPR) is one of the classical problems in the field of object recognition. Its application is very crucial in the automation of transportation system since it helps to recognize a vehicle identity, which information is stored in the license plate. LPR usually consists of three major phases: pre-processing, license plate localization, optical character recognition (OCR). Despite being classical, its implementation faced with much more complex problems in the real scenario, mainly caused by natural lighting, specular reflection in the vehicle body, and object with high saturation intensity. This paper proposed an improved LPR algorithm based on modified Horizontal-Vertical Edge Projection. It is used to detect and localize the region of interest by doing such analysis along the horizontal and vertical projection of the image. Related works proved that the modified Horizontal-Vertical Edge Projection is the simplest method to be implemented, yet very effective against Indonesian license plate. However, its performance gets reduced when specular reflection is occurred in the sample image. Therefore, morphological operations are utilized in the pre-processing phase to reduce the effects of specular reflection and high saturation intensity, while preserving the needed information to detect region of interest. Based on the experimental results, our proposed algorithm shows an improvement compared with the previous research and successfully detect 71 license plates in 80 image samples which results in 88.75% accuracy. Key words: license plate recognition, morphology operation, image processing, computer vision License Plate Recognition (LPR) adalah salah satu permasalahan klasik dalam bidang pengenalan objek. Aplikasinya sangat penting bagi otomatisasi sistem transportasi dikarenakan dapat membantu untuk mengenali identitas kendaraan, berdasarkan informasi yang tersimpan dalam plat nomor. LPR biasanya terdiri atas tiga fase besar: pre-processing, license plate localization, optical character recognition (OCR). Meskipun klasik, implementasinya dihadapkan dengan berbagai permasalahan yang sangat kompleks pada skenario yang sesungguhnya, umumnya disebabkan oleh faktor cahaya, refleksi spekular pada bagian kendaraan, serta objek dengan intensitas saturasi tinggi. Jurnal ini mengusulkan sebuah pengembangan dari algoritma LPR berdasarkan metode Horizontal-Vertical Edge Projection yang telah dimodifikasi. Metode tersebut digunakan untuk mendeteksi dan melakukan lokalisasi terhadap region of interest melalui suatu analisis di sepanjang proyeksi horizontal dan vertikal pada gambar. Penelitian terkait membuktikan bahwa metode Horizontal-Vertical Edge Projection merupakan metode yang implementasinya sangat sederhana, namun begitu efektif terhadap plat nomor kendaraan Indonesia. Meskipun demikian, performanya dapat menurun secara drastis apabila sampel gambar mengandung refleksi spekular. Oleh karena itu, operasi morfologi dimanfaatkan sedemikian rupa pada fase pre-processing untuk mengurangi efek dari refleksi spekular dan intensitas saturasi tinggi, namun tetap mempertahankan berbagai informasi yang diperlukan untuk mendeteksi region of interest. Berdasarkan hasil eksperimen, algoritma yang kami usulkan memperlihatkan sebuah kemajuan bila dibandingkan dengan penelitian sebelumnya dan berhasil mendeteksi 71 plat nomor dari 80 sampel gambar sehingga didapati nilai akurasi sebesar 88.75%. Kata kunci: pengenalan plat nomor, pengolahan citra, operasi morfologi, computer vision

Actions (login required)

View Item