| 摘要 |
PURPOSE:To efficiently eliminate the noise component included in a composite image by clipping the image within a prescribed range when no orthogonal transformation coefficient of the image exists in such a range and obtaining the final image by applying prescribed processing to an obtained correction image. CONSTITUTION:After encoded data inputted from a reception terminal 501 is decoded by a decoder part 503, a quantization representative value for each orthogonal transformation coefficient is reproduced by an inverse quantization part 504, and inverse orthogonal transformation is applied to it in an inverse orthogonal transform part 505, then, decoded image data 506 can be obtained. The image data 506 is inputted to a clipping part 509 via an orthogonal transform part 508. While, a detecting part 510 finds the maximum value and the minimum value of the coefficient of the quantization representative value, and a clipping range is set by a setting part 511, and the clipping part 509 substitutes the orthogonal transform coefficient in the range. and the inverse orthogonal transformation is applied to it by an inverse orthogonal transform part 513, then, a first correction image 514 can be obtained. The image data receives orthogonal transformation 516 via an LPF part 515, and the orthogonal transform coefficients are corrected by a clipping part 517, and inverse orthogonal transformation 518 is applied to it, then, a second correction image 519 can be obtained. Similarly, an n-th correction image 520 can be outputted by applying such processing repeatedly for prescribed number of times. |
