The experimental results show that the average PSNR of the proposed algorithm can reach 33.5889 dB in IMDB face database, which is higher than that of other methods and the recovery of face image has a better visual effect.
对IMDB人脸库进行了实验对比，结果显示，新方法得到了33.588 9 dB的平均峰值信噪比，高于Baker方法和传统的插值算法，显示出了更好的视觉效果。
The result of simulation experiments shows that IAIR could increase PSNR of the reconstructed images effectively in case of a high packet-loss rate.
计算机仿真的实验结果表明，在IP信道分组丢失率较高的情况下，交互式自适应帧内刷新算法可以有效的提高恢复视频的PSNR。
Comparing with the bit-plane coding in MPEG-4 FGS, our CABIC scheme further achieves a PSNR gain of 0.5~1.0dB.
比起MPEG-4的位元层编码，我们的内文适应性位元层编码更进一步的达到0.5~1dB的PSNR改善。
The method in the paper is based on Embedded Block Coding with Optimized Truncation, which is done by adaptive distortion tracing. The decoder does not need to reversely lift the coefficients and works normally. More importantly, this method remains all the features of EBCOT such as resolution progressive, PSNR progressive and good robust for error bits spread.
从而克服了传统方法因为增强系数与图像复杂度不匹配带来的ROI与BG的PSNR质量不协调的问题，该算法的解码器不需要知道该图像是否存在ROI，不需要反提升过程，完全正常解码即可，而且该方法保留了EBCOT的优良特性。
The obtained fusion image is evaluated in subjective and objective factors. Experiments show that this method can yield fusion images with better visual quality and better objective evaluation evaluation criteria in entropy, mean square error, PSNR, similarity and so on.
对得到的融合图像进行了主客观评价和对比，实验结果表明，本文方法得到的融合图像，在熵、均方误差、峰值信噪比和相似度等客观指标上都优于其他方法。
In the end, this model was tested for the image denoising. It was also compared with contourlet HMT and wavelet thresholding using PSNR, NMSE and visual quality.
最后，将此分布应用于图像去噪，就PSNR、NMSE和视觉质量这三方面的评价指标与contourlet HMT和小波阈值法进行了比较。
Two image quality evaluation of the function. Including psnr PSNR, it is mean-square error nmse other. Images can be used for de-noising and compression of image quality evaluation.
详细说明：两个图像质量评价函数，psnr峰值信噪比，正则均方误差nmse，能可用于去噪图像和压缩图像的质量评价，很有用啊。
Our experiments show that DSNR of different images in the same scene changes congruously with PSNR, thus can be used to evaluate the quality of decompressed images and image sequences.
实验证明，在相同场景中DSNR能较好地反映PSNR的变化，因而可以用于对同一图象的不同解压缩图象以及图象序列的评价。
For example, for Barbara image 512×512, using 8192 most significant coefficients, the PSNR of reconstruct image outperforms contourlet's by 0.68 dB.
例如对于512×512的Barbara图，在保持8192个重要系数时，重建图像的峰值信噪比较采用Contourlet变换时提高0.68 dB。
Experimental results showed that the new estimator recovers line-type structures well and produces higher PSNR than other existing algorithms.
实验结果表明，估计子不仅能得到较满意的线形结构暠，而且能获得比现有方法更高的信噪比。
In the thesis, the fractal theory is introduced into the frequency domain from the spatial domain, and a DCT domain fractal image coding scheme is proposed. DDFIC scheme can improve the quality of the reconstructed images efficiently resulting in the PSNR of more than 31dB. The PSNR is more than 6dB higher than the traditional. A fast algorithm for DCT based on matrix decomposition is proposed. The number of its multiplication is the lowest at present. A new affine transform is proposed. Compared with the traditional fractal image coding scheme, the PSNR achieved by NAT scheme increases nearly 4 dB while maintaining the compression ratio. The contractivity and the convergence of fractal image coding are analyzed. Several theorems and lemmas about the contractivity are presented in the thesis, and they are much helpful in choosing fractal parameters. In NAT scheme, the least mean square method is adapted. It makes the block matching faster and more accurate and therefore reduces the computing complexity and speeds up the compression. A sequence image coding scheme based on extended self-similarity fractal is proposed in the thesis.
DDFIC方案非常有效地提高了恢复图象的质量，其恢复图象的信噪比大都在31dB以上，比传统分形编码方案提高了6dB以上，这对分形编码来说是一个质的飞跃；提出一种基于矩阵分解的DCT快速算法，在目前来说，其乘法次数是最少的；提出一种新的仿射变换，与传统分形编码方案相比，NAT方案在压缩比基本不变的情况下，使信噪比提高了近4dB；在国际上首次对分形编码的紧缩性和收敛性进行了定量分析，给出了几个定理和引理，使编码方案在选择参数时有据可查；在NAT方案中，采用最小二乘法求解最佳分形变换参数，使寻找匹配过程既快又准，避免了传统方案在寻找最佳参数的巨大运算量，其编码速度明显加快；在序列图象编码中，提出了扩展自相似分形序列图象编码方案。
By using the Human Vision System properties into the intra and inter frame processing, the proposed method can overcome the traditional Peak Signal-to-noise Ratio in the application of video quality evaluation.
通过在视频帧内图像和帧间图像的处理过程中引入人眼视觉系统的主要特性，克服传统 PSNR 算法在序列质量检测应用方面的缺陷。
Then a method to detect scene change for intra frames is proposed. Based on the new characters of H.264/AVC and the results of presented scene change detection, the block based temporal concealment with boundary match criteria and spatial error concealment scheme is adaptively adopted for inter and intra frames. The proposed methodology can make use of the available redundancy in bit streams as possible.
实验结果表明，所提方法在错误隐藏的重构图像质量在主观感觉和客观PSNR定量上，都比通常的方法有明显的提高；与传统错误隐藏方法相比，本文的基于场景切换检测的错误隐藏机制鲁棒性高，更能适应的H.264/AVC更为复杂的多种应用场合。
Finally, an image compression scheme, including DSP for multimedia process、2D-DWT transformer、video camera and network transmission, is designed for the rotary kiln alumina production. Compression experiments are carried out to compress and reconstruct the flame image in rotary keln under different PSNR condition at various spatial and temporal resolutions. The coal particles region can be identified accurately at very low bitrate. Experiment results show the validity of the scheme proposed for flame image compression in rotary keln.
通过回转窑火焰图像的压缩进行了仿真实验，实现了黑把子图像在不同空间分辨率、时间分辨率和不同PSNR条件下的压缩与重构，完成了在极低码率下黑把子的准确识别，说明了该系统可有效应用于回转窑火焰图像的压缩。
The compression results on many standard test images show that, using the same coding method, LBT based one performs better than DWT based, with higher PSNR and less distortion.
对大量标准测试图像的压缩结果表明，采用同样的编码方法，基于LBT压缩后图像的峰值信噪比普遍高于DWT，且主观质量也有明显改善。
Experimental results show that our algorithm can handle frame loss within different movements intensity and obtain higher PSNR and better visual effects.
实验表明，该方法能较好地解决不同运动剧烈程度下的丢帧问题，使解码图像获得较高的PSNR和较好的视觉效果。
The experimental results show that the proposed method obviously outperforms the neighborhood threshold shrinkage method in the ratio of peak signal to noise.
通过实验证明，该方法取得了比邻域阈值萎缩法更高的PSNR值，同时对图像细节进行增强，得到了更佳的视觉效果。
After advantages and disadvantages of conventional algorithms for the motion compensation having been analyzed, a new fast search algorithm based on combining temporal dependence of video with spatial dependence of video is proposed, the emulated results shows that the coding efficiency and the PSNR of the decoded is improved greatly, meanwhile fast searching speed can be obtained.
在分析了常用的几种运动补偿算法的优缺点后，提出了一种基于时间相关性和空间相关性的快速搜索算法，仿真结果表明，在获得较快的搜索速度的同时，编码效率和峰值信噪比得到大大提高。
The method can enhance the accuracy of estimation, weaken the harmful effect on the motion edge by global smoothness constraint in Horn algorithm, and improve the PSNR.
该算法能够提高光流估计的精确性，降低Horn-Schunck法中全局平滑约束对运动边界的影响，提高峰值信噪比。
Using this method, accuracy of estimation can be enhanced, the harmful effect on the motion edge by global smoothness constraint in Horn algorithm can be weakened and the PSNR of estimation can be improved.
这种算法能够较精确进行光流运动估计，减轻Horn方法中全局平滑约束对运动边界的影响，提高估计的峰值信噪比。