해상환경에서의 표적식별 성능 향상을 위한 적외선영상 화질개선 기법
Infrared image enhancement method to improve target recognition in marine environments
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
국방 연구개발 분야에서 적외선영상은 표적의 탐지, 추적 시스템에 주로 사용된다. 하지만 적외선영상은 히스토그램 명암비가 한쪽으로 치우친다는 특성이 있기 때문에 목표물에 대한 추적 시스템에 효과적으로 활용하기 위해서는 명암비 개선이 필수적이다. 이러한 특성을 보완하기 위해 많은 기법들이 연구되었지만, 해상환경의 적외선영상과 같이, 영상의 히스토그램 분포가 배경성분의 밝기값에 집중될 경우, 표적정보가 소실되거나, 표적영역에서의 과도한 밝기 변화로인한 과포화 현상이 여전히 발생한다는 문제점이 있다. 따라서 본 논문에서는 해상환경에서의 표적식별 성능 향상을 위한 화질개선 기법을 제안하고자 한다. 이를 위해, 먼저 변형된 Retinex알고리듬을 기반으로 영상의 배경성분을 제거한뒤, 그 결과값을 이용하여 Plateau Histogram Equalization의 임계치를 적응적으로 설정한다. 실험 수행 결과, 기존 적외선영상 화질개선 기법과 대비하여 표적의 명암비 향상, 과포화 현상 및 표적의 세부정보 손실의 문제점이 개선됨을 확인/ 검증하였다.
Abstract
In defense R&D, infrared images are commonly used for target detection and tracking systems. However, infrared images have a weakness in that their distribution of histograms for dynamic range leans toward one side. Therefore, improving the quality of images for robust performance and improved system effectiveness is necessary. Extensive research has been conducted to solve this problem; however, two problems remain with respect to infrared images in marine environments in which the histogram distribution is concentrated in the gray level of background components. One problem concerns lost target information caused by the failure to set a proper threshold. The other concerns over-enhancement due to the excessive change in brightness in the target region. For this reason, this paper proposes an effective image enhancement method for infrared images for target recognition in a marine environment. The first step of the proposed method is to remove the background components of the image by using a modified Retinex algorithm. The next step is to determine the adaptive threshold of plateau histogram equalization to increase the target dynamic range using the data derived from the first step. Experiments were performed with marine infrared images, and the results indicate that our proposed method shows better enhancement performance compared with other conventional methods.
참고문헌 (14)
- R. C. Gonzalez, "Digital Image Processing" (2006), Prentice-Hall, Inc.
- V. E. Vichers, "Plateau equalization algorithm for real-time display of high-quality infrared imagery", Optical Engineering, 35(7), pp.1921~1926 (1996)
- B. Wang, "A real-time contrast enhancement algorithm for infrared images based on plateau histogram", Infrared Physics and Technology, 48, pp.77~82 (2006)
- A. Raju, "Histogram equalization based mean self-adaptive plateau histogram equalization for brightness preserving and contrast enhancement", The 3rd World Conference on Applied Sciences, Engineering and Technology, pp.232~239 (2014)
- C. H. Ooi, "Bi-histogram equalization with a plateau limit for digital image enhancement", IEEE Transactions on Consumer Electronics, 55(4), pp.2072~2080 (2009)
- A. S. Parihar, "Contrast enhancement using entropy-based dynamic sub-histogram equalization", The Institution of Engineering and Technology Image Processing, 10(11), pp.799~808 (2016)
- Z. Yao, "Quadrants histogram equalization with a clipping limit for image enhancement", 2016 8th International Conference on Wireless Communications & Signal Processing, pp.1~5 (2016)
- A. Paul, "Plateau limit-based tri-histogram equalization for image enhancement", The Institution of Engineering and Technology Image Processing, 12(9), pp.1617~1625 (2018)
- D. J. Jobson, "Properties and performance of a center", surround retinex, 1997, pp.6(3), 451~462
- E. Land, "Lightness and retinex theory", Journal of the Optical Society of America, 61(1), pp.1~11 (1971)
- 이원석, "Retinex 처리에 기반한 적외선 열상 이미지의 화질 개선", 전자공학회논문지 - IE, 48(2), pp.32~39 (2011), 대한전자공학회
- Z. U. Rahman, "Multi-scale retinex for color image enhancement", Proceedings on 3rd IEEE International Conference on Image Processing, pp.1003~1006 (1996)
- T. H. Lee, "Adaptive windowing for sharpness enhancement and halo reduction", The Institute of Electronics of Electronics Engineers of Korea, pp.893~894 (2008)[DOI]
- L. Zhuang, "Image enhancement via subimage histogram equalization based on mean and variance", Hindawi Computational Intelligence and Neuroscience, 2017, pp.1~12 (2017)