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
차량 로고 감지 기술은 지능형 교통 시스템 적용에 대한 연구 방향 중 하나입니다. 이는 번호판과 오토바이 유형을 기반으로 한 감지 기술의 중요한 확장입니다. 차량 로고의 특징은 독특함, 가독성, 다양성입니다. 그러므로 이론과 응용에 있어서 철저한 연구가 중요하다. 객체 감지와 관련된 일부 작업이 있지만 대부분은 다양한 장면에 대한 실시간 감지를 달성할 수 없습니다. 한편, 단일 단계의 일부 실시간 감지 방법은 작은 크기의 객체 감지에서 성능이 좋지 않았습니다. 학습 샘플이 부족한 문제를 해결하기 위해 본 논문의 작업은 차량 로고(VLD-45-S) 데이터 구성, 다단계 사전 학습, 다단계 예측, 특징 융합을 통해 개선되었습니다. 얕은 레이어가 있는 더 깊은 층, 경계 상자의 차원 클러스터링, 다중 규모 탐지 훈련 사이에 있습니다. 이 기사는 속도 유지를 기반으로 차량 로고의 감지 정밀도를 향상시킵니다. 실제 장면의 감지 모델 일반화 및 간섭 방지 기능은 데이터 강화를 통해 최적화됩니다. 실험 결과, 작은 크기의 물체에 대해 감지 알고리즘의 정확도와 속도가 향상되는 것으로 나타났습니다.
Junxing ZHANG
Dalian Minzu University
Shuo YANG
Kyushu Institute of Technology
Chunjuan BO
Dalian Minzu University
Huimin LU
Kyushu Institute of 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.
부
Junxing ZHANG, Shuo YANG, Chunjuan BO, Huimin LU, "Single Stage Vehicle Logo Detector Based on Multi-Scale Prediction" in IEICE TRANSACTIONS on Information,
vol. E103-D, no. 10, pp. 2188-2198, October 2020, doi: 10.1587/transinf.2020EDP7088.
Abstract: Vehicle logo detection technology is one of the research directions in the application of intelligent transportation systems. It is an important extension of detection technology based on license plates and motorcycle types. A vehicle logo is characterized by uniqueness, conspicuousness, and diversity. Therefore, thorough research is important in theory and application. Although there are some related works for object detection, most of them cannot achieve real-time detection for different scenes. Meanwhile, some real-time detection methods of single-stage have performed poorly in the object detection of small sizes. In order to solve the problem that the training samples are scarce, our work in this paper is improved by constructing the data of a vehicle logo (VLD-45-S), multi-stage pre-training, multi-scale prediction, feature fusion between deeper with shallow layer, dimension clustering of the bounding box, and multi-scale detection training. On the basis of keeping speed, this article improves the detection precision of the vehicle logo. The generalization of the detection model and anti-interference capability in real scenes are optimized by data enrichment. Experimental results show that the accuracy and speed of the detection algorithm are improved for the object of small sizes.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2020EDP7088/_p
부
@ARTICLE{e103-d_10_2188,
author={Junxing ZHANG, Shuo YANG, Chunjuan BO, Huimin LU, },
journal={IEICE TRANSACTIONS on Information},
title={Single Stage Vehicle Logo Detector Based on Multi-Scale Prediction},
year={2020},
volume={E103-D},
number={10},
pages={2188-2198},
abstract={Vehicle logo detection technology is one of the research directions in the application of intelligent transportation systems. It is an important extension of detection technology based on license plates and motorcycle types. A vehicle logo is characterized by uniqueness, conspicuousness, and diversity. Therefore, thorough research is important in theory and application. Although there are some related works for object detection, most of them cannot achieve real-time detection for different scenes. Meanwhile, some real-time detection methods of single-stage have performed poorly in the object detection of small sizes. In order to solve the problem that the training samples are scarce, our work in this paper is improved by constructing the data of a vehicle logo (VLD-45-S), multi-stage pre-training, multi-scale prediction, feature fusion between deeper with shallow layer, dimension clustering of the bounding box, and multi-scale detection training. On the basis of keeping speed, this article improves the detection precision of the vehicle logo. The generalization of the detection model and anti-interference capability in real scenes are optimized by data enrichment. Experimental results show that the accuracy and speed of the detection algorithm are improved for the object of small sizes.},
keywords={},
doi={10.1587/transinf.2020EDP7088},
ISSN={1745-1361},
month={October},}
부
TY - JOUR
TI - Single Stage Vehicle Logo Detector Based on Multi-Scale Prediction
T2 - IEICE TRANSACTIONS on Information
SP - 2188
EP - 2198
AU - Junxing ZHANG
AU - Shuo YANG
AU - Chunjuan BO
AU - Huimin LU
PY - 2020
DO - 10.1587/transinf.2020EDP7088
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2020
AB - Vehicle logo detection technology is one of the research directions in the application of intelligent transportation systems. It is an important extension of detection technology based on license plates and motorcycle types. A vehicle logo is characterized by uniqueness, conspicuousness, and diversity. Therefore, thorough research is important in theory and application. Although there are some related works for object detection, most of them cannot achieve real-time detection for different scenes. Meanwhile, some real-time detection methods of single-stage have performed poorly in the object detection of small sizes. In order to solve the problem that the training samples are scarce, our work in this paper is improved by constructing the data of a vehicle logo (VLD-45-S), multi-stage pre-training, multi-scale prediction, feature fusion between deeper with shallow layer, dimension clustering of the bounding box, and multi-scale detection training. On the basis of keeping speed, this article improves the detection precision of the vehicle logo. The generalization of the detection model and anti-interference capability in real scenes are optimized by data enrichment. Experimental results show that the accuracy and speed of the detection algorithm are improved for the object of small sizes.
ER -