The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Keadaan jerebu akan menjejaskan kualiti pengecaman plat lesen dan mengurangkan prestasi algoritma pemprosesan visual. Bagi meningkatkan kualiti gambar jerebu, algoritma pengecaman plat lesen berdasarkan cuaca jerebu dicadangkan dalam kertas kerja ini. Algoritma dalam kertas ini terutamanya terdiri daripada dua bahagian: Bahagian pertama ialah penyahjerebuan imej MPGAN, yang menggunakan rangkaian musuh generatif untuk menyahjerebu imej, dan menggabungkan lilitan berbilang skala dan kehilangan persepsi. Konvolusi berbilang skala adalah kondusif untuk pengekstrakan ciri yang lebih baik. Kehilangan persepsi menebus kelemahan bahawa ralat min kuasa dua (MSE) banyak dipengaruhi oleh outlier; bahagian kedua ialah mengenali plat lesen, mula-mula kita menggunakan YOLOv3 untuk mencari plat lesen, rangkaian STN membetulkan plat lesen, dan akhirnya memasuki rangkaian LPRNet yang dipertingkatkan untuk mendapatkan maklumat plat lesen. Keputusan eksperimen menunjukkan bahawa model dehazing yang dicadangkan dalam kertas ini mencapai keputusan yang baik, dan penunjuk penilaian PSNR dan SSIM adalah lebih baik daripada algoritma perwakilan lain. Selepas membandingkan algoritma pengecaman plat lesen dengan algoritma LPRNet, kadar ketepatan purata boleh mencapai 93.9%.
Weiguo ZHANG
Xi'an University of Science and Technology
Jiaqi LU
Xi'an University of Science and Technology
Jing ZHANG
Xi'an University of Science and Technology
Xuewen LI
Xi'an University of Science and Technology
Qi ZHAO
Xi'an University of Science and Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Weiguo ZHANG, Jiaqi LU, Jing ZHANG, Xuewen LI, Qi ZHAO, "Research on the Algorithm of License Plate Recognition Based on MPGAN Haze Weather" in IEICE TRANSACTIONS on Information,
vol. E105-D, no. 5, pp. 1085-1093, May 2022, doi: 10.1587/transinf.2021EDP7178.
Abstract: The haze situation will seriously affect the quality of license plate recognition and reduce the performance of the visual processing algorithm. In order to improve the quality of haze pictures, a license plate recognition algorithm based on haze weather is proposed in this paper. The algorithm in this paper mainly consists of two parts: The first part is MPGAN image dehazing, which uses a generative adversarial network to dehaze the image, and combines multi-scale convolution and perceptual loss. Multi-scale convolution is conducive to better feature extraction. The perceptual loss makes up for the shortcoming that the mean square error (MSE) is greatly affected by outliers; the second part is to recognize the license plate, first we use YOLOv3 to locate the license plate, the STN network corrects the license plate, and finally enters the improved LPRNet network to get license plate information. Experimental results show that the dehazing model proposed in this paper achieves good results, and the evaluation indicators PSNR and SSIM are better than other representative algorithms. After comparing the license plate recognition algorithm with the LPRNet algorithm, the average accuracy rate can reach 93.9%.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2021EDP7178/_p
Salinan
@ARTICLE{e105-d_5_1085,
author={Weiguo ZHANG, Jiaqi LU, Jing ZHANG, Xuewen LI, Qi ZHAO, },
journal={IEICE TRANSACTIONS on Information},
title={Research on the Algorithm of License Plate Recognition Based on MPGAN Haze Weather},
year={2022},
volume={E105-D},
number={5},
pages={1085-1093},
abstract={The haze situation will seriously affect the quality of license plate recognition and reduce the performance of the visual processing algorithm. In order to improve the quality of haze pictures, a license plate recognition algorithm based on haze weather is proposed in this paper. The algorithm in this paper mainly consists of two parts: The first part is MPGAN image dehazing, which uses a generative adversarial network to dehaze the image, and combines multi-scale convolution and perceptual loss. Multi-scale convolution is conducive to better feature extraction. The perceptual loss makes up for the shortcoming that the mean square error (MSE) is greatly affected by outliers; the second part is to recognize the license plate, first we use YOLOv3 to locate the license plate, the STN network corrects the license plate, and finally enters the improved LPRNet network to get license plate information. Experimental results show that the dehazing model proposed in this paper achieves good results, and the evaluation indicators PSNR and SSIM are better than other representative algorithms. After comparing the license plate recognition algorithm with the LPRNet algorithm, the average accuracy rate can reach 93.9%.},
keywords={},
doi={10.1587/transinf.2021EDP7178},
ISSN={1745-1361},
month={May},}
Salinan
TY - JOUR
TI - Research on the Algorithm of License Plate Recognition Based on MPGAN Haze Weather
T2 - IEICE TRANSACTIONS on Information
SP - 1085
EP - 1093
AU - Weiguo ZHANG
AU - Jiaqi LU
AU - Jing ZHANG
AU - Xuewen LI
AU - Qi ZHAO
PY - 2022
DO - 10.1587/transinf.2021EDP7178
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E105-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2022
AB - The haze situation will seriously affect the quality of license plate recognition and reduce the performance of the visual processing algorithm. In order to improve the quality of haze pictures, a license plate recognition algorithm based on haze weather is proposed in this paper. The algorithm in this paper mainly consists of two parts: The first part is MPGAN image dehazing, which uses a generative adversarial network to dehaze the image, and combines multi-scale convolution and perceptual loss. Multi-scale convolution is conducive to better feature extraction. The perceptual loss makes up for the shortcoming that the mean square error (MSE) is greatly affected by outliers; the second part is to recognize the license plate, first we use YOLOv3 to locate the license plate, the STN network corrects the license plate, and finally enters the improved LPRNet network to get license plate information. Experimental results show that the dehazing model proposed in this paper achieves good results, and the evaluation indicators PSNR and SSIM are better than other representative algorithms. After comparing the license plate recognition algorithm with the LPRNet algorithm, the average accuracy rate can reach 93.9%.
ER -