Search

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.

Deep convolutional neural networks have been used extensively in recent image processing research, exhibiting drastically improved performance. In this study, we apply convolutional neural networks to color filter array demosaicking, which plays an essential role in single-sensor digital cameras. Contrary to conventional convolutional neural network-based demosaicking models, the proposed model does not require any initial interpolation step for mosaicked input images, which increases the computational complexity. Using a mosaicked image as input, the proposed model is trained in an end-to-end manner to generate demosaicked images outputs. Many deep neural networks experience vanishing-gradient problem, which makes models hard to be trained. To solve this problem, we apply residual learning and densely connected convolutional neural network. Moreover, we apply block-wise convolutional neural networks to consider local features. Finally, we apply a sub-pixel interpolation layer to generate demosaicked output images more efficiently and accurately. Experimental results show that our proposed model outperforms conventional solutions and state-of-the-art models.

The salient object detection (SOD) technology predicts which object will attract the attention of an observer surveying a particular scene. Most state-of-the-art SOD methods are top-down mechanisms that apply fully convolutional networks (FCNs) of various structures to RGB images, extract features from them, and train a network. However, owing to the variety of factors that affect visual saliency, securing sufficient features from a single color space is difficult. Therefore, in this paper, we propose a multi-color space network (MCSNet) to detect salient objects using various saliency cues. First, the images were converted to HSV and grayscale color spaces to obtain saliency cues other than those provided by RGB color information. Each saliency cue was fed into two parallel VGG backbone networks to extract features. Contextual information was obtained from the extracted features using atrous spatial pyramid pooling (ASPP). The features obtained from both paths were passed through the attention module, and channel and spatial features were highlighted. Finally, the final saliency map was generated using a step-by-step residual refinement module (RRM). Furthermore, the network was trained with a bidirectional loss to supervise saliency detection results. Experiments on five public benchmark datasets showed that our proposed network achieved superior performance in terms of both subjective results and objective metrics.

In this paper, we introduce a weighted sum-based color filter array interpolation method using Taylor series cubic approximation. We use a high-order approximation to predict accurate pixel values during the reconstruction of quincunx grid-sampled green channels, and perform a weighted average-based interpolation in a large local window. We also perform prediction utilizing a color difference model in a small local window to generate additional green values. By applying the weighted sum method to the predicted green values in two local windows, green channel can be enhanced. The remaining color components, namely, the rectangular grid-sampled red and blue channels, are interpolated utilizing the weighted average method of the color difference model. Additionally, we propose a post-processing method that removes the zipper artifacts generated during the interpolation process. Experimental results demonstrate that the proposed color filter array interpolation system outperforms existing algorithms in terms of both objective and subjective performance.

In this paper, we introduce a new color filter array interpolation based on edge map prediction using the Bayesian theorem. The edge information obtained at the position of the green component sampled in the quincunx grid using the Bayer pattern is used to predict the edge of the red and blue component positions distributed in the rectangular grid. In this process, the edge distribution of the entire image, local area, and nearest neighbors is analyzed to determine the smooth/edge characteristics stochastically. If the pixels at the red and blue positions are determined to be edge pixels, the edge direction is inherited from the edge pixels of the local region. By combining a new decision technique with directional weighted interpolation in the green plane reconstruction process, the interpolation accuracy can be improved compared with the existing demosaicking algorithms. After interpolating the green channel, the color difference between the interpolated green channel and the red/blue channel is used to refine the reconstructed green plane. The other color planes can then be interpolated using this refined green plane. Experimental results show that our algorithm improves both the objective and subjective image qualities compared with the conventional state-of-the-art demosaicking algorithms.

In this paper, we propose a new demosaicking system that uses post-processing technique to remove zipper artifacts that are primarily caused by improper interpolation in a quincunx-sampled green plane. We use a high-order approximation to predict accurate pixel values in the weighted average-based green channel interpolation. We also perform prediction using a color difference model in the red and blue channel interpolation process. As a well-designed demosaicking system too cannot prevent the occurrence of ON-OFF pattern artifacts and the resulting false colors, we propose post-processing for zipper artifact detection and removal to improve the demosaicked image. After performing the just noticeable difference modelbased zipper artifact detection in the luminance image of the demosaicked image, the clamping-based zipper artifact removal process is performed in the red, green, and blue planes. Experimental results demonstrate that the proposed demosaicking system improves both the objective and subjective image qualities compared to the conventional state-of-the-art demosaicking algorithms.

With the recent development of video compression methods, video transmission on traditional devices and video distribution using networks has increased in various devices such as drones, IP cameras, and small IoT devices. As a result, the demand for encryption techniques such as MPEG-DASH for transmitting streams over networks is increasing. These video stream security methods guarantee stream confidentiality. However, they do not hide the fact that the encrypted stream is being transmitted over the network. Considering that sniffing attacks can analyze the entropy of the stream and scan huge amounts of traffic on the network, to solve this problem, the deception method is required, which appears unencrypted but a confidential stream. In this paper, we propose the new deception method that utilizes standard NAL unit rules of video codec, where the unpromised device shows the cover video and the promised device shows the secret video for deceptive security. This method allows a low encryption cost and the stream to dodge entropy-based sniffing scan attacks. The proposed stream shows that successful decoding using five standard decoders and processing performance was 61% faster than the conventional encryption method in the test signal conformance set. In addition, a network encrypted stream scan method the HEDGE showed classification results that our stream is similar to a compressed video.

The use of error-correcting codes (ECCs) is essential for designing reliable digital communication systems. Usually, most systems correct errors under cooperative environments. If receivers do not know interleaver parameters, they must first find out them to decode. In this paper, a blind interleaver parameters estimation method is proposed using the Kolmogorov–Smirnov (K–S) test. We exploit the fact that rank distributions of square matrices of linear codes differ from those of random sequences owing to the linear dependence of linear codes. We use the K–S test to make decision whether two groups are extracted from the same distribution. The K–S test value is used as a measure to find the most different rank distribution for the blind interleaver parameters estimation. In addition to control false alarm rates, multinomial distribution is used to calculate the probability that the most different rank distribution will occur. By exploiting those, we can estimate the interleaver period with relatively low complexity. Experimental results show that the proposed algorithm outperforms previous methods regardless of the bit error rate.

Ultrasound (US) imaging can examine human bodies of various ages; however, in the process of obtaining a US image, speckle noise is generated. The speckle noise inhibits physicians from accurately examining lesions; thus, a speckle noise removal method is essential technology. To enhance speckle noise elimination, we propose a novel algorithm using the characteristics of speckle noise and filtering methods based on speckle reducing anisotropic diffusion (SRAD) filtering, discrete wavelet transform (DWT) using symmetry characteristics, weighted guided image filtering (WGIF), and gradient domain guided image filtering (GDGIF). The SRAD filter is exploited as a preprocessing filter because it can be directly applied to a medical US image containing speckle noise without a log-compression. The wavelet domain has the advantage of suppressing the additive noise. Therefore, a homomorphic transformation is utilized to convert the multiplicative noise into additive noise. After two-level DWT decomposition is applied, to suppress the residual noise of an SRAD filtered image, GDGIF and WGIF are exploited to reduce noise from seven high-frequency sub-band images and one low-frequency sub-band image, respectively. Finally, a noise-free image is attained through inverse DWT and an exponential transform. The proposed algorithm exhibits excellent speckle noise elimination and edge conservation as compared with conventional denoising methods.

Synthetic aperture radar (SAR) images map Earth’s surface at high resolution, regardless of the weather conditions or sunshine phenomena. Therefore, SAR images have applications in various fields. Speckle noise, which has the characteristic of multiplicative noise, degrades the image quality of SAR images, which causes information loss. This study proposes a speckle noise reduction algorithm while using the speckle reducing anisotropic diffusion (SRAD) filter, discrete wavelet transform (DWT), soft threshold, improved guided filter (IGF), and guided filter (GF), with the aim of removing speckle noise. First, the SRAD filter is applied to the SAR images, and a logarithmic transform is used to convert multiplicative noise in the resulting SRAD image into additive noise. A two-level DWT is used to divide the resulting SRAD image into one low-frequency and six high-frequency sub-band images. To remove the additive noise and preserve edge information, horizontal and vertical sub-band images employ the soft threshold; the diagonal sub-band images employ the IGF; while, the low- frequency sub-band image removes additive noise using the GF. The experiments used both standard and real SAR images. The experimental results reveal that the proposed method, in comparison to state-of-the art methods, obtains excellent speckle noise removal, while preserving the edges and maintaining low computational complexity.