(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Traffic sign and license plate detection based on saliency, meanshift, and mathematical morphology

Traffic sign and license plate detection based on saliency, meanshift, and mathematical morphology / Cheong Wei Sheik
Sebuah model pengesanan objek yang terdiri daripada septrum ketonjolan, peruasan anjakan purata, dan anggaran bentuk morfologi telah dicadangkan dalam kajian ini. Kaedah pengiraan ketonjolan yang lebih baik berdasarkan perhatian penglihatan manusia telah diperkenalkan. Septrum ketonjolan adalah berdasarkan prinsip operasi de-konvolusi dalam domain log-spektrum, dan pengiraannya cepat dengan hanya parameter tunggal untuk ditala. Selain itu, septrum ketonjolan mempamerkan ciri keteguhan warna di bawah pelbagai pencahayaan, dengan skim warna biasa RGB boleh digunakan untuk imej warna. Bagi meningkatkan lagi prestasi model yang dicadangkan, anjakan purata tanpa parametrik dan kaedah Otsu telah digunakan untuk meruaskan objek daripada sekitarnya. Selain itu, kaedah faktor bentuk yang mudah berdasarkan morfologi matematik diperkenalkan untuk mengenal pasti objek yang diruas dengan mengukur bentuknya. Untuk menilai keberkesanan dan kesesuaian kaedah yang dicadangkan, dua masalah di sektor pengangkutan, iaitu, pengesanan tanda isyarat lalu-lintas dan plat lesen kenderaan dikaji secara terperinci. Berdasarkan dua set data daripada umum dan dikumpul secara tempatan, kaedah yang dicadangkan menunjukkan keseimbangan yang baik antara ketepatan dan kelajuan. Keputusan simulasi menunjukkan bahawa ia adalah tujuh kali lebih cepat daripada teknik perihalan bentuk dalam pengesanan tanda isyarat lalu-lintas, dan mengambil masa kurang daripada 0.6 saat dalam pengesanan plat lesen kenderaan berbanding degan kaedah pemadanan pencontoh dan pembelajaran mesin. Kajian ini menunjukkan kegunaan kaedah pengesanan objek dalam satu kerangka kerja bersepadu untuk kedua-dua masalah pegesanan tanda isyarat lalu-lintas dan plat lesen kenderaan, oleh itu, menyumbang ke arah kemajuan dalam sistem pengangkutan pintar. _______________________________________________________________________________________________________ An object detection model that consists of cepstrum saliency, mean-shift segmentation, and morphological shape estimation is proposed in this research. An improved computational saliency method based on human visual attention is introduced. Cepstrum saliency is based on the principles of de-convolution in the log-spectrum domain, and is computationally fast with only single parameter to tune. Moreover, cepstrum saliency exhibits color consistency under various illuminations, where the normalized RGB color scheme can be used for color images. To further enhance the proposed object detection model, non-parametric mean-shift and Otsu’s method are utilized for figure-ground segmentation. Besides that, simple shape factors based on mathematical morphology are introduced to identify the segmented objects by measuring shapes. To evaluate the effectiveness and applicability of the proposed method, two problems in the transportation section, i.e., traffic sign and license plate detection, were studied in detail. Based on two publicly available and locally collected data sets, the proposed detection method demonstrates a good equipoise between accuracy and speed. The simulation results indicate that it is seven times faster than shape descriptors in traffic sign detection, and has an average of less than 0.6 s in license plate detection as compared with template matching and machine learning methods. The findings indicate the usefulness of the proposed object detection method in providing a unified framework for both traffic sign and license plate detection problems; therefore contributing towards advancement in intelligent transportation systems.
Contributor(s):
Cheong Wei Sheik - Author
Primary Item Type:
Thesis
Identifiers:
Accession Number : 
Language:
English
Subject Keywords:
object detection; computational saliency; human visual
First presented to the public:
9/1/2015
Original Publication Date:
8/27/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 145
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-08-29 15:33:30.138
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Traffic sign and license plate detection based on saliency, meanshift, and mathematical morphology1 2018-08-29 15:33:30.138