Showing papers on "Quantization (image processing) published in 1981"
TL;DR: The Hough transform was improved by O'Gorman and Clowes by taking into account the gradient direction and the resulting scatter of the peaks can be reduced by using a weighting function in the transform.
196 citations
01 Jan 1981
TL;DR: In this article, the effect of limited word-length on the performance of fast discrete cosine transform (FDCT) processors is analyzed using mathematical analysis, and the theoretical results are verified and extended for further detailed simulation studies.
Abstract: For hardware realization of the fast discrete cosine transform (FDCT) processor, as the parameters can only be specified by a finite number of binary digits (bits), the computational accuracy is governed by the precision by which the parameters are represented. To measure the effect of limited word-length, mathematical analysis is used initially to develop theoretical effects of the sensitive parameters on the processing accuracy. By simulation, the theoretical results are verified, and the results are then extended for further detailed simulation studies. Finally, the simulation results are used to develop general guidelines for optimizing the hardware design of any size FDCT or fast inverse DCT (FIDCT) processor.
6 citations
TL;DR: Two modifications to the constant area quantization (CAQ) video bandwidth compression scheme are described which together reduce the mean square error (MSE) in the reconstructed image by up to 70 percent.
Abstract: Two modifications to the constant area quantization (CAQ) video bandwidth compression scheme are described which together reduce the mean square error (MSE) in the reconstructed image by up to 70 percent The first modification is merely a change in the CAQ algorithm requiring no additional hardware, whereas the second modification exploits line-to-line redundancies and necessitates the inclusion of one line of memory
4 citations
01 Apr 1981
TL;DR: A non-adaptive scheme for encoding an image sequence is described that yields better results than any previous comparable interframe image coding method and uses a full image DFT and an optimal quantization algorithm.
Abstract: A non-adaptive scheme for encoding an image sequence is described that yields better results than any previous comparable interframe image coding method. The scheme uses a hybrid of two-dimensional transform coding for the spatial dimension and DPCM for the time dimension similar to some previous works [e.g., 1]. Unlike the previous efforts, which use discrete cosine transforms (DCT's) of image subblocks and suboptimal quantization, we utilize a full image DFT and an optimal quantization algorithm. The results with the standard sixteen frame sequence of Walter Cronkite are visually superior to previous transform coding efforts, especially at rates lower than 1 bit/pixel/ frame. The signal-to-noise ratios are higher than those of nonadaptive hybrid DCT/DPCM in [1] by 2.6 to 10.5 dB depending on rate.
3 citations
TL;DR: Hybrid sampling is used in both two dimensional transform and hybrid (transform-DPCM) processing of standard digitized images of the GIRL and MOONSCAPE as mentioned in this paper.
05 Apr 1981
TL;DR: Several discrete transform methods, including the KL and FFT, have been compared in order to determine their relative utility in an image processing environment and the effect of sectioning an image into sub-blocks has been investigated.
Abstract: Several discrete transform methods, including the KL and FFT, have been compared in order to determine their relative utility in an image processing environment. Computational requirements ( both memory size and execution time), mean-squared error, and also subjective factors have been includedinthestudy . In addition, t he effect of sectioning an image into sub-blocks has been investigated. R esults should be useful in the design and implementation of an image data compression system.