SITE MAP
주요사업
- 시험분석
- 품질검사
- 제품안전·인증·검증
-
제품안전(KC)
- 연구개발
- 기업지원
기술정보
직/편성물 염색
A Comparative Study of a CRT Colorimetric Prediction Model by Neural Networks and the Models by Conventional Method
- 출판일1999.03
- 저자
- 서지사항
- 등록일 2016.11.02
- 조회수 283
Researchers developed a CRT colorimetric prediction model using a set of artificial neural networks (ANN) in which a set of back propagation (BP) neural networks performs a nonlinear conversion between red, green, and blue (RGB) values and tristimulus values (X.Y.Z). Theoretical comparisons of the neural network based model with conventional CRT colorimetric prediction models indicated that the new model would resolve such troubles as constant channel chromaticity, gun or channel independence, and screen background effects. An experiment trained a set of four layer BP neural networks and performed the color conversion function in a computer controlled CRT system. The ANN model provided satisfactory prediction results. 12 refs.
