KOREASCHOLAR

과적/적재 불량 차량 추적을 위한 이미지 유사도 비교 방법 연구 Study on Image Similarity for Tracking Overloaded and Poor Loaded Vehicles

이경수, 윤득선, 임병철
  • 언어KOR
  • URLhttp://db.koreascholar.com/Article/Detail/444704
PDF 원문보기
한국기계기술학회지 (韓國機械技術學會誌)
제27권 제4호 (2025.08)
pp.614-619
한국기계기술학회 (Korean Society of Mechanical Technology)
초록

Overloaded and improperly loaded trucks cause serious road hazards, such as rollovers and cargo falls. Although automatic enforcement methods are being studied, they face challenges in accuracy and legal application. Thus, a technology for direct tracking and enforcement is needed. This study uses EfficientNet to extract features of vehicles and license plates, and applies cosine similarity to identify the same vehicle. Comparisons were divided into “same vehicle” and “similar vehicle,” with a threshold-based method and five classification types. Results showed that the average similarity of the same vehicle group was 0.11 higher than that of the similar vehicle group. The accuracy of correctly identifying the same vehicle was 84.54%. Integrating OCR or LPR is expected to further improve tracking performance.

키워드
화물차과적적재 불량추적영상 인식딥러닝이피션트넷유사도 TruckOverloadPoor loadTrackingImage RecognitionDeep LearningEfficientNetSimilarity
목차
Abstract
1. 서 론
2. 연구 내용
    2.1 기존 추적 방법
    2.2 데이터 수집
    2.3 연구 방법
    2.4 이미지 유사도 비교 방법
3. 결 과
4. 결 론
후 기
References
저자
  • 이경수(Researcher, AI Semiconductor R&D Center. Korea Automotive Technology Institute (Katech)) | K. S. Lee Corresponding author
  • 윤득선(Executive Principal Researcher, AI Semiconductor R&D Center. Korea Automotive Technology Institute (Katech)) | D. S. Yun
  • 임병철(Executive Principal Researcher, AI Semiconductor R&D Center. Korea Automotive Technology Institute (Katech)) | B. C. Yim