http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Signal Recovery from Wavelet Threshold Crossings
Kikuchi, Hisakazu,Nakashizuka, Makoto,Ishii, Ikuo,Makino, Hideo 대한전자공학회 1994 ISPACS:Intelligent Signal Processing and Communica Vol.1 No.1
This paper presents signal reconstruction from wavelet threshold-crossings by using convex projection. We define the wavelet threshold crossings as the irregular sampling of a discrete dyadic wavelet transform. The reconstruction is done by the projection between a convex which characterized by the threshold-crossings and a linear space of all possible wavelet transform. The wavelet threshold-crossings convey the same information as the quantized wavelet transform. The reconstruction precision of the threshold-crossing is superior to the inverse wavelet transform of the quantized wavelet transform. Image reconstruction from the wavelet threshold crossings is demonstrated.
Hisakazu Kikuchi,Shogo Muramatsu 대한전자공학회 2009 ITC-CSCC :International Technical Conference on Ci Vol.2009 No.7
A lossless image compression is presented as a bit-plane coding that exploits a scheme of bit modeling by the expectation values of pixel values at a decoder. The algorithm is simple and various functionality extensions are implemented including selective tile partitioning, progressive transmission, ROI transmission, accuracy scalability, and others. The mean squared error between the original image and a decoded image at any progression level is known prior to encoding. The proposed bit-plane codec is competitive with JPEG-LS and JPEG 2000 in the lossless compression of 8-bit grayscale and 24-bit color images.
Junghyeun Hwang,Hisakazu Kikuchi,Shogo Muramatsu,Jaeho Shin 대한전자공학회 2009 ITC-CSCC :International Technical Conference on Ci Vol.2009 No.7
In this paper, we propose an error diffusion halftoning algorithm for grayscale image using blue noise spectra. The algorithm is search for the error diffusion filter coefficients in all gray levels (i.e. intensity levels) by optimizing the total error between the obtained Fourier spectrum and the designed blue noise spectrum for blue noise mask (BNM). At each gray level, BNM is designed so that the pattern has the construction of binary patterns, and also its power spectrum has blue noise (high frequency) characteristics in the frequency domain which are the human visual system (HVS). Experimental results show that error diffusion filter is obtained to design BNM according to various parameters of the binary pattern power spectrum manipulation algorithm (BIPPSMA).