Showing papers on "Peak signal-to-noise ratio published in 2001"

A method and system for estimating objective quality of compressed video data

Abstract: The present invention relates to a method and system for estimating the quality of encoded video data without referring to the source data. The system is configured to decode compressed video data using an MPEG/H.263 decoder to produce decompressed video data. The decoded (16) data is subject to a discrete cosine transform (DCT) to produce a set of DCT coefficients of the decompressed video data is determined to be intra-coded. Meanwhile, the quantization matrix including a quantization step size for each block of the decompressedvideo data are extracted (24). Following this, statistical properties of the DCT coefficients are estimated (26) from the decoded video. Then, an average quantization error for both AC and DC coefficients is determined (28) using information regarding the quantization and the statistical properties. Finally, this estimated quantization error is used for in computing (28) a peak signal to noise ratio (PSNR).

Resolution enhancement of monochrome and color video using motion compensation

Abstract: We propose an iterative algorithm for enhancing the resolution of monochrome and color image sequences. Various approaches toward motion estimation are investigated and compared. Improving the spatial resolution of an image sequence critically depends upon the accuracy of the motion estimator. The problem is complicated by the fact that the motion field is prone to significant errors since the original high-resolution images are not available. Improved motion estimates may be obtained by using a more robust and accurate motion estimator, such as a pel-recursive scheme instead of block matching, in processing color image sequences, there is the added advantage of having more flexibility in how the final motion estimates are obtained, and further improvement in the accuracy of the motion field is therefore possible. This is because there are three different intensity fields (channels) conveying the same motion information. In this paper, the choice of which motion estimator to use versus how the final estimates are obtained is weighed to see which issue is more critical in improving the estimated high-resolution sequences. Toward this end, an iterative algorithm is proposed, and two sets of experiments are presented. First, several different experiments using the same motion estimator but three different data fusion approaches to merge the individual motion fields were performed. Second, estimated high-resolution images using the block matching estimator were compared to those obtained by employing a pel-recursive scheme. Experiments were performed on a real color image sequence, and performance was measured by the peak signal to noise ratio (PSNR).

Unequal error protection applied to JPEG image transmission using turbo codes

Abstract: An investigation of unequal error protection (UEP) methods applied to JPEG image transmission using turbo codes is presented. The JPEG image is partitioned into two groups, i.e., DC components and AC components according to their respective sensitivity to channel noise. The highly sensitive DC components are better protected with a lower coding rate, while the less sensitive AC components use a higher coding rate. Simulation results are given to demonstrate how the UEP schemes outperforms the equal error protection (EEP) scheme in terms of bit error rate (BER) and peak signal to noise ratio (PSNR).

Robust invisible watermarking of volume data using the 3D DCT

Abstract: Proposes a novel watermarking algorithm for 3D volume data based on the spread-spectrum communication technique, which is invisible and robust. "Invisible" means that the 2D rendered image of this watermarked volume is perceptually indistinguishable from that of the original volume. "Robust" watermarking implies that the watermark is resistant to most intentional or unintentional attacks. We have implemented the algorithm in a software package and have conducted experiments showing that the watermark is invisible in the volume-rendered 2D images. This is further confirmed by computing the signal-to-noise ratio (SNR) and the peak signal-to-noise ratio (PSNR) of the 3D watermarked volume data. We addressed different attacks, putting more emphasis on those attacks that are most commonly performed by attackers. The experiments show that the watermarking scheme is robust.

A General Watermarking Framework for Optimal Energy Estimation

Abstract: Digital watermarking has been proposed as a solution to the problem of copyright protection of multimedia documents in network environments Most of the current image watermarking schemes is based on unitary transforms, such as discrete cosine transform, discrete Fourier transform and discrete wavelet transform, etc In general an effective watermarking scheme should satisfy the following two requirements: imperceptibility and robustness It can be seen that the two requirements are in fact contrary to each other For a particular watermarking scheme, we have to consider the trade off between imperceptibility and robustness Thus how to determine the optimal watermark energy or capacity inserted into the original documents should be solved according to some evaluation criteria (such as the commonly used Peak Signal to Noise Ratio PSNR and Signal to Noise Ratio SNR) In this paper we addressed the issue of how to evaluate the energy and capacity of the inserted watermark We define the two dimensional discrete unitary transform Under the assumption of the transform, the digital watermarking is done in the unitary transform domain and the additive watermark embedding model is defined (which is adopted by many current watermarking schemes) Thus, given the permitted threshold of visual quality metrics (especially the most popular measurement PSNR and SNR), we can determine the best energy or capacity to be added to the original image The estimation formula is given upon the assumption of unitary transforms and additive embedding models Finally, we discuss two general watermark types (normal Gaussian white noise and binary noise) in our experiments to illustrate the correctness of the theoretical results From the experimental results, we can see that the theoretical values are consistent with the experiments

Improvements to SPHIT for lossy image coding

Abstract: Proposes five measures to improve the SPIHT (set partitioning in hierarchical trees) algorithm for lossy image coding. First, a new type of tree with virtual root is introduced to hold more wavelet coefficients. Second, an extra matrix is used to speed up the judgement of the significance of trees. Third, a pre-processing is done to smooth the coefficients before SPIHT encoding. Fourth, some predictable bits are omitted from the encoder output by rearranging the coding procedure. Finally, the quantisation is offset from the middle-point according to the statistics. Our experiments show that these improvements increase the PSNR (peak signal to noise ratio) by up to 5 dB at very low rates, and the average improvement at 0.2-1 bit/pixel is about 0.5 dB for the standard test images used. Despite the performance gain, the computational complexity is also reduced.

Hierarchical video motion estimation using a neural network

Abstract: In our previous papers, we have presented a neural network-based motion estimation technique that is applicable to purely translational as well as affine movements. In this paper we present a further advance, namely a hierarchical version which, while capitalizing on the merits of the previous technique, achieves some additional advantages. These are: (a) the ability to deal with very large motion action sequences; and (b) a reduction in the number of candidate macroblocks for the neural network stage and correspondingly, the number of computations needed. In this hierarchical technique, the first stage consists of one of the familiar block matching techniques, such as the LBM or FS, over very coarse pixels, to define a suitable candidate neighborhood. In the second, and final, stage a few of the neighbors of the candidate found in first stage are used in the computation of affinities via the neural network to synthesize the source macroblock. It produces 4 to 11 dB improvement in peak signal-to-noise ratio over the single-candidate methods, thus creating the potential for improved compression.

Study of moving image quality evaluation base on mpeg-2 system

Abstract: The evaluation of image quality includes subjective evaluation and objective test.The subjective evaluation is complex and its results always change,and the development of objective evaluation is necessary.The perfection of the objective test depends on the comprehension of TV system matched by human visual characteristics.Combining the character of MPEG 2 system and human visual characteristics,the authors of this paper studied deeply two distortions about the evaluation of moving inage:three dimensional weighted SNR (WSNR) and three dimensional weighted block distortion (WBSNR).The experimental results show that compared with the physical peak signal to noise ratio (PSNR),correlation of the objective test parameter in this paper with the subjective evaluation the result is greatly improved.

Robust 3-D subband video coder

Abstract: We explain our results on development of a robust video coder, The coder consists of 3D subband filtering in the first stage, an optimized scalar quantization (SQ) of lowest tempo-spatial frequency subband, and a pyramid vector quantization (PVQ) for high frequency subbands. To increase robustness to noise productive enumeration is used for PVQ. Also with phase scrambling the distribution of the lowest temporal-spatial band is changed to Gaussian which results in a better SQ coding gain for this band. The output indices of SQ were sent using a folded binary code to improve noise robustness. Results show that at high bit-error rate (5/spl times/10/sup -3/) the coder peak signal to noise ratio drops a maximum 5.3 dB.

A New Image Encoding Algorithm Based on Wavelet Zero-Tree Coding Scheme

Abstract: In this paper, a new low bit rate image compression algorithm based on wavelet zero tree image coding scheme is put forward. The coefficients in the horizontal and vertical orientation sub images of all scales are reordered by the means of the transform because there are many big energy coefficients. The target of doing so is to produce new tree structure, to add the number of zero tree. And the significant wavelet coefficients in these sub images are concentrated in the low frequency area of the corresponding subbands. So the energy of these sub images is concentrated in the area of the corresponding left hand, and the con centration property of the new trees is good, and it can improve the encoding efficiency effectively. It also can improve the transmission rate of the significant coefficients. The results of the reconstruction images are good. And bit allocation is optimized and the peak signal to noise ratio ( PSNR ) of the reconstructed image is improved compared to embedded zero tree wavelet encoding(EZW) algorithm at the same bit ratio.

Robust 3-dimensional subband video coder based on dithered pyramid vector quantizer

Abstract: We present our results on the development of a robust video coder. The system consists of 3-dimensional subband filtering, optimized scalar quantization of the lowest temporal-spatial frequency subband and modified pyramid vector quantization for high frequency subbands. To improve the visual quality of the system at low bit-rates, dithering is used for the pyramid vector quantizers and its effectiveness is proved by a pair-comparison subjective test. To increase robustness to noise, productive enumeration is used for the pyramid vector quantizers. Also, with phase scrambling, the distribution of the lowest temporal-spatial band is changed to Gaussian which results in a better scalar quantization coding gain for this band and, by using a folded binary code, less sensitivity to noise. Results shows that at a high bit-error rate (5/spl times/10/sup -3/), the coder in average loses only 4 dB in peak signal-to-noise ratio.

Application of pulse compression with a Doppler shift using neural network

Abstract: Common biphase pulse compression techniques were not very robustness against Doppler distortions, and the peak signal-to-noise ratio of these methods was not very high. Kwan and Lee (1993) presented a pulse compression approach by using a neural network. A 40 dB peak signal-to-noise ratio can be achieved. Based on Kwan and Lee's method, we present an improved BP (backpropagation) algorithm. The robustness against Doppler distortions and the output peak signal-to-noise ratio of the improved algorithm are examined. The improved method get a 50 dB output peak signal-to-noise ratio, about 10 dB higher than Kwan and Lee's method. The improved method also has a better Doppler resistance ability.

Two-level unequal error protection scheme in image transmission system using multilevel codes in Rayleigh fading channel

Abstract: We propose a novel unequal error protection (UEP) scheme with two levels for image transmission using multilevel codes (MLC). By providing the best protection for the most important data, the final recovered image quality is remarkably improved both in visual effect and in peak-signal-to-noise power ratio (PSNR) performance.

Apparatus for controlling bit rate of mpeg encoder

Abstract: PURPOSE: An apparatus for controlling a bit rate of an MPEG(Motion Picture Expert Group) encoder is provided to improve a compression rate of a motion picture by using PSNR(Peak Signal to Noise Ratio) information to control a bit rate CONSTITUTION: A motion estimation and compensation portion(17) extracts a motion vector and obtains a predicting value of a macro block of an input picture A subtracter(10) calculates a difference value between the predicted block and an input picture block A DCT portion(11) converts the difference value by DCT A quantization portion(12) quantizes the converted data A variable length coding portion(13) codes the quantized data An inverse quantization portion(15) quantizes inversely the coded data A bit rate control portion(14) controls an output of the quantization portion A PSNR calculation portion(18) compares PSNR of each picture to each other

Image coding using adaptive vector quantization of wavelet coefficients

Abstract: In this paper we propose a subband image compression by using wavelet transform to split original images. Each of subband images is then quantized by an adaptive vector quantization with dynamic bit allocation based on advantage of nature of wavelet coefficients. The energy of each subband image, except the lowest frequency subband image will not be quantized, will be sorted from minimum to maximum. Energy of each subband image is calculated to allocate bits not over the desired bit rate. The accumulation of energy from these subband images will be divided into 4 groups. First two lower energy groups will be encoded with 256 and 1 6 code vectors for 1 6 pixels block size in accordance with energy ratio. Others will be encoded with 256 code vectors for 4 and 16 pixels block size. Based on the given bit rate, the total dynamical bit rate of each group is calculated. If the total dynamical bit rate in the group is less or more than the given bit, it will thenbe adjusted based on the energy of subband image in only the same group. The remaining of energy from higher energy group will be carried to lower. The experiments are shown that the resulting images from the proposed method can be clearly improved by Peak Signal to Noise Ratio (PSNR) of 36.30 16, MSE =15.2377, 1 .03 125 bpps.

Parameters and filters for low bit rate wavelet packet compression of magnetic resonance images

Abstract: We present an analysis of the characteristics of different filters for the compression of magnetic resonance images. Compression rates were 33:1 and 50:1. We compare the performance among different types of wavelets presented in the literature and provide quantitative (percentage of energy retained, peak signal to noise ratio) and qualitative (analysis by a group of seven experts) data to support our conclusions. Different types of coiflets, symlets and biorthogonal wavelets are analyzed, and we conclude that for the images under study (T1 weighed images in three planes), the best results are provided by the biorthogonal spline (Daubechies) wavelet 2,6. Several explanations for these results are mentioned.

A concealment scheme for H.263 coded video transported over the Internet using the RTPUDPIP protocol

Abstract: Packet video with real-time constraints, limited bandwidth, multicast distribution, and using the RTPUDPIP protocol must take into account packet losses (with a large payload size) due to network congestion An effective solution for packet loss is to perform concealment at the receiver In this paper, we propose a concealment scheme for the 'I' frame missing odd or even slices that address the above problems The proposed concealment scheme uses the previous frame macroblock (MB) or the MB motion vector (MV) with a fast motion estimation algorithm, located at the same spatial location as the missing MB The decision between the MB or MB MV with a fast motion estimation algorithm is determined by the amount of motion between the decoded upper and lower 1 pixel wide boundary (if available) of the missing MB and the MB located in same spatial location in the previous frame The proposed scheme produced similar peak signal to noise ratio (PSNR) results while the Central Processing Unit (CPU) time is reduced by a factor of 30 when compared to the full search concealment technique