• 제목/요약/키워드: piece-wise linear curve

검색결과 5건 처리시간 0.018초

Piece-wise linear estimation of mechanical properties of materials with neural networks

  • Shin, Inho
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국제학술편); KOEX, Seoul; 19-21 Oct. 1992
    • /
    • pp.181-186
    • /
    • 1992
  • Many real-world problems are concerned with estimation rather than classification. This paper presents an adaptive technique to estimate the mechanical properties of materials from acoustoultrasonic waveforms. This is done by adapting a piece-wise linear approximation technique to a multi-layered neural network architecture. The piece-wise linear approximation network (PWLAN) finds a set of connected hyperplanes that fit all input vectors as close as possible. A corresponding architecture requires only one hidden layer to estimate any curve as an output pattern. A learning rule for PWLAN is developed and applied to the acousto-ultrasonic data. The efficiency of the PWLAN is compared with that of classical backpropagation network which uses generalized delta rule as a learning algorithm.

  • PDF

햇빛 아래에서 향상된 시인성을 위한 Piece-wise Linear Enhancement Curves 기반 영상 개선 (Image Enhancement based on Piece-wise Linear Enhancement Curves for Improved Visibility under Sunlight)

  • 이준민;송병철
    • 방송공학회논문지
    • /
    • 제27권5호
    • /
    • pp.812-815
    • /
    • 2022
  • 햇빛 아래에서 디지털 기기에 표시되는 영상은 일반적으로 원본 영상보다 어둡게 인식되어 시인성이 저하된다. 더 나은 시인성을 위해서, 주변광에 적응적인 전역 휘도 보상 혹은 톤 매핑이 필요하다. 하지만 기존의 기법들은 색차 보상에 한계가 존재하고, 무거운 계산 비용 때문에 실제 환경에서 활용하는데 어려움이 존재한다. 이를 해결하기 위해 본 논문에서는 휘도와 색차를 모두 보상하는 Piece-wise Linear Enhancement Curves (PLECs) 기반 영상 개선 기법을 제안한다. 이때, PLECs은 딥러닝을 통해 회귀 되며, lookup table 형식으로 구현되어 실시간 동작이 가능하다. 실험 결과 제안 방법이 낮은 계산 비용으로 원본 영상 대비 더 나은 시인성을 가짐을 보인다

Implementation of Adaptive Shading Correction System Supporting Multi-Resolution for Camera

  • Ha, Joo-Young;Song, Jin-Geun;Im, Jeong-Uk;Min, Kyoung-Joong;Kang, Bong-Soon
    • 융합신호처리학회 학술대회논문집
    • /
    • 한국신호처리시스템학회 2006년도 하계 학술대회 논문집
    • /
    • pp.25-28
    • /
    • 2006
  • In this paper, we say the shading correction system supporting multi-resolution for camera. The shading effect is caused by non-uniform illumination, non-uniform camera sensitivity, or even dirt and dust on glass (lens) surfaces. In general this shading effect is undesirable [1]. Eliminating it is frequently necessary for subsequent processing and especially when quantitative microscopy is the fine goal. The proposed system is available on thirty nine kinds of image resolutions scanned by interlaced and progressive type. Moreover, the system is using various continuous quadratic equations instead of using the piece-wise linear curve which is composed of multiple line segments. Finally, the system could correct the correct effect without discontinuity in any image resolution. The proposed system is also experimentally demonstrated with Xilinx Virtex FPGA XCV2000E- 6BG5560 and the TV set.

  • PDF

VLSI Implementation of Adaptive Shading Correction System Supporting Multi-Resolution for Mobile Camera

  • Ha, Joo-Young;Lee, Sung-Mok;Jang, Won-Woo;Yang, Hoon-Gee;Kang, Bong-Soon
    • 한국통신학회논문지
    • /
    • 제31권12C호
    • /
    • pp.1201-1207
    • /
    • 2006
  • In this paper, we say the adaptive shading correction system supporting multi-resolution for mobile camera. The shading effect is caused by non-uniform illumination, non-uniform camera sensitivity, or even dirt and dust on glass (lens) surfaces. In general this shading effect is undesirable [1-3]. Eliminating it is frequently necessary for subsequent processing and especially when quantitative microscopy is the fine goal. The proposed system is available on thirty nine kinds of image resolutions scanned by interlaced and progressive type. Moreover, the system is using forty kinds of continuous quadratic equations instead of using the piece-wise linear curve which is composed of multiple line segments. Finally, the system could correct the shading effect without discontinuity in any image resolution. The proposed system is implemented in VLSI with cell library based on Hynix $0.25{\mu}m$ CMOS technology.