Compression artifact reduction based on total variation regularization method for MPEG-2

Goto, Tomio; Kato, Yasuyuki; Hirano, Satoshi; Sakurai, Masaru; Nguyen, Truong Q.

doi:10.1109/tce.2011.5735510

Cited by 19 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Experiments are conducted on 10 standard images compressed at high compression ratio. Results show that on an average, the MSSIM values of the images restored using the proposed method improved by 1.10%, 2.49%, 1.14% and 1.89% compared to the methods given in [12], [13], [14] and [15] respectively.…”

Section: A Results Of Weighted Averaging With Negative Like Imagementioning

confidence: 93%

“…In order to show the effect of the technique more clearly, the same portion of the image which has been displayed in The PSNR values and the mean square error values are normally used to measure the quality of the reconstructed images. But, PSNR or MSE does not correlate well with the human perception of quality [15] because of the following reasons. Firstly, the digital pixel values, on which the MSE is typically computed, may not exactly represent the light stimulus entering the eye.…”

Section: A Results Of Weighted Averaging With Negative Like Imagementioning

confidence: 99%

See 1 more Smart Citation

Deblocking of Gray Scale Images and Videos Using 3D Spatial Filtering

Chacko¹,

Jayakumar²

2014

ijeei

View full text Add to dashboard Cite

Digital images are compressed at high compression ratios for storage and communication purposes and are thereby affected by visual blocking artifact. Various filtering techniques are available for reducing the visual blocking artifact of compressed images. In this paper, a 3D spatial filter which uses a negative like image of the image to be deblocked to reduce the compression artifact of the compressed image. The Mean Structural Similarity Index Measure (MSSIM) and Peak Signal to Noise Ratio (PSNR) are used in this paper to assess the quality of the restored images. The results obtained show that these techniques effectively remove the compression artifact from the compressed digital images.

show abstract

Section: A Results Of Weighted Averaging With Negative Like Imagementioning

confidence: 93%

Section: A Results Of Weighted Averaging With Negative Like Imagementioning

confidence: 99%

Deblocking of Gray Scale Images and Videos Using 3D Spatial Filtering

Chacko¹,

Jayakumar²

2014

ijeei

View full text Add to dashboard Cite

show abstract

“…Some methods have been proposed to restore JPEG images. Liu et al [6] proposed sparsity-based dual-domain DCT, and Goto et al [7] reduced compression artifacts using total variation regularization. Nosratinia [8] proposed enhancement of the compressed image by re-application of JPEG compression, and Jancsary et al [9] proposed a method using regression tree fields.…”

Section: Introductionmentioning

confidence: 99%

Local Excitation Network for Restoring a JPEG-Compressed Image

2019

IEEE Access

View full text Add to dashboard Cite

Joint photographic experts group (JPEG) compression is lossy compression, and degradation of image quality worsens at high compression ratios. Therefore, a reconstruction process is required for a visually pleasant image. In this paper, we propose an end-to-end deep learning architecture for restoring JPEG images with high compression ratios. The proposed architecture changes a core principle of the squeeze and excitation network for low-level vision tasks where pixel-level accuracy is important. Instead of extracting global features, our network extracts locally embedded features and fine-tunes each feature value by using depthwise convolution. To reduce the computational complexity and parameters with large receptive fields, we use a combination of the recursive structure and feature map down-and up-scaling processes. We also propose a compact version of the proposed model by decreasing the number of filters and simplifying the network, which has about one-twentieth of the parameters of the baseline model. Experimental results reveal that our network outperforms conventional networks quantitatively, and the restored images are clear with sharp edges and smooth blocking boundaries. Furthermore, the compact model shows higher objective results while maintaining a low number of parameters. In addition, at a high compression ratio, the overall information, including details in the blocks, are lost owing to high quantization errors. We apply a generative adversarial network structure to restore these highly damaged blocks, and the results reveal that the image produced has details similar to those of the ground truth.INDEX TERMS Convolutional neural network, JPEG image restoration, generative adversarial network.

show abstract

“…Various priors have been exploited [9,4,23,3]. Jung et al attempt to reconstruct the intrinsic layer via sparse representation, which, however, requires the compression ratio is known and the dictionary is well-learned [13].…”

Section: Introductionmentioning

confidence: 99%