http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : 김범재(Beomjae Kim) (검색결과 1 건)
- 무료
- 기관 내 무료
- 유료
-
이미지 전처리를 이용한 저조도 환경에서의 이미지 인식률 향상 방법
안우경(Wookyung An),김기찬(Ki Chan Kim),김범재(Beomjae Kim) 한국자동차공학회 2022 한국자동차공학회 학술대회 및 전시회 Vol.2022 No.11
Many vehicle manufacturers are trying to develop an autonomous driving car, also known as a self-driving car, hence, automotive software for Advanced Driver Assistance System (ADAS) is getting more complex than ever before. This is because the autonomous driving system can shift a new paradigm for car driving environments (e.g., watching TV while driving a car). Accordingly, object detection (OD) is very important to achieve the ADAS framework since interpreting the visual information (e.g., object classification and its properties) closes the relationship with morphological image analysis and understanding of the classification as well as properties of the object within the image sequence. Recently, the proliferation of deep-learning-based OD frameworks, the advanced of the inexpensive camera attached to the vehicle, and the increasing need for automated video analysis. However, some limited environments (e.g., low-illumination) give arise to several challenging factors: vague classification of the ground truth image data. To overcome the limited scenario, we introduced a novel preprocessing method to enhance the images obtained from the low illumination environments. Furthermore, we compared the result of Precision, Recall, and F1 score from the previous DL-based OD algorithm and the proposed method. The proposed method outperforms 15.87% in recall, and 15.93% F1 score than the previous object detection method. We believe that the proposed method will provide important guidance for improving the camera-based future ADAS framework whether high-illumination or low-illumination environments.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
