JPEG Lossy Research Articles

Influence of lossy JPEG compression on measurement uncertainty for luminance-based area measurements

Zusammenfassung Es kann aus verschiedenen Gründen nötig sein, günstige Consumer-Kameras für Labormessungen einzusetzen. Diese haben jedoch häufig nur einen Videostream mit verlustbehafteter Komprimierung. Wir zeigen in diesem Artikel, am Beispiel des Motion-JPEG (MJPEG)-Codecs, dass dies für helligkeitsbasierte Flächenmessungen keine Einschränkungen bedeuten muss. Zentrale Rolle spielt hier der Hell/Dunkel-Übergang, welcher die Fläche eingrenzt. Der Effekt von JPEG auf solche Kanten wird jedoch in der Literatur nur für Kantenbreiten von wenigen Pixeln untersucht. Wir implementieren den wichtigsten Schritt der JPEG-Komprimierung in Matlab, und vergleichen zwei Messreihen zu Flächenmessungen aus jeweils 1.800 Bildern einer Pupille und eines Metalloches. Dabei werden der Qualitätsfaktor des JPEG-Codecs und die Breite des Hell/Dunkel-Überganges variiert. Als Ergebnis liegt die mittlere quadratische Abweichung der Flächenmaße nach Komprimierung für alle untersuchten Breiten (zwischen 4 und 22,5 Pixeln) und alle ausgewerten Qualitätsfaktoren (zwischen 30 und 90) unterhalb der Standardabweichung der Flächenmessung des Metalloches. Im Detail ergibt sich schließlich eine weitere Abhängigkeit von der Steilheit des Hell/Dunkel-Überganges.

An exhaustive measurement of re-sampling detection in lossy compressed images using deep learning approach

The detection of resampling in digital images is critical for image authentication, but performance can be challenging when dealing with lossy compression. This study proposes an efficient feature extraction technique for detecting resampling (i.e., tampering) in post-JPEG compressed images. Our approach combines compression clues with resampling clues and feeds them to various traditional machine learning (ML) algorithms such as logistic regression, K-nearest neighbours (K-NN), support vector machine (SVM), decision tree (DT), and random forest (RF) to detect and classify doctored images in the re-compression scenario. We propose and evaluate feed-forward deep neural networks (DNN) and 1D convolutional neural networks (CNN) based on evaluation parameters such as accuracy, recall, precision, and F1 score, comparing them with the aforementioned traditional ML algorithms. Our results show that the RF and one-dimensional (1D) CNN are the most efficient models for this task. Furthermore, the 1D CNN outperforms the state-of-the-art techniques, particularly in the most challenging case of downscaling in lossy JPEG compressed images. Our proposed method demonstrates promising results for resampling detection in post-JPEG compressed images, which can be helpful in various image authentication applications.

Robust image steganography against lossy JPEG compression based on embedding domain selection and adaptive error correction

Transmitting images for communication on social networks has become routine, which is helpful for covert communication. The traditional steganography algorithm is unable to successfully convey secret information since the social network channel will perform lossy operations on images, such as JPEG compression. Previous studies tried to solve this problem by enhancing the robustness or making the cover adapt to the channel processing. In this study, we proposed a robust image steganography method against lossy JPEG compression based on embedding domain selection and adaptive error correction. To improve anti-steganalysis performance, the embedding domain is selected adaptively. To increase robustness and lessen the impact on anti-steganalysis performance, the error correction capacity of the error correction code is adaptively adjusted to eliminate redundancy. The experimental results show that the proposed method achieves better anti-steganalysis and robustness.

Forensic analysis of manipulation chains: A deep residual network for detecting JPEG-manipulation-JPEG

Due to the proliferation of image editing software, the problem of identifying doctored photos has risen to prominence in information forensics. Even though existing image processing history detection techniques could provide reasonably acceptable outcomes on publicly available databases, most of them generally fail to accomplish this when the doctored images are lossy compressed, which is commonly accomplished in social media. To tackle this issue, a dual-stream residual network (ReMReNet) using ResNet as a backbone is proposed for JPEG-resistant image operator chain detection. ReMReNet consists of noise residual extraction (NRE) and compression feature extraction (CFE) streams for better results against JPEG compression. The CFE stream is trained on discrete cosine transform (DCT) residual coefficients by producing clues left by JPEG-agent images with generic JPEG compression artefacts. In addition, the NRE stream is utilized to extract noise residuals to suppress the image content and enhance the manipulation clues. Hence, the domain transformation between uncompressed and post-JPEG compressed images is reduced by training the backbone network with a domain adaption approach, which helps forensic investigators pinpoint the exact location of the tampering. The proposed framework outperforms other state-of-the-art algorithms in image multiple manipulations and operator chain detection, notably for lossy JPEG compression, as shown by extensive experimental findings on numerous heterogeneous datasets.

Robust Image Steganography: Hiding Messages in Frequency Coefficients

Steganography is a technique that hides secret messages into a public multimedia object without raising suspicion from third parties. However, most existing works cannot provide good robustness against lossy JPEG compression while maintaining a relatively large embedding capacity. This paper presents an end-to-end robust steganography system based on the invertible neural network (INN). Instead of hiding in the spatial domain, our method directly hides secret messages into the discrete cosine transform (DCT) coefficients of the cover image, which significantly improves the robustness and anti-steganalysis security. A mutual information loss is first proposed to constrain the flow of information in INN. Besides, a two-way fusion module (TWFM) is implemented, utilizing spatial and DCT domain features as auxiliary information to facilitate message extraction. These two designs aid in recovering secret messages from the DCT coefficients losslessly. Experimental results demonstrate that our method yields significantly lower error rates than other existing hiding methods. For example, our method achieves reliable extraction with 0 error rate for 1 bit per pixel (bpp) embedding payload; and under the JPEG compression with quality factor QF=10, the error rate of our method is about 22% lower than the state-of-the-art robust image hiding methods, which demonstrates remarkable robustness against JPEG compression.

Effects of compressed speckle image on digital image correlation for vibration measurement

Digital image correlation (DIC) can accurately measure subpixel displacement field based on speckle images. However, in vibration measurement, the extensive high-resolution images captured by DIC systems impose significant data transmission and storage requirements on hardware. A framework combining image compression techniques with DIC methods is proposed. The framework effectively reduces the memory footprint of images while preserving the primary speckle features with subpixel errors. The DIC Challenge 2.0 image sets are employed in simulation to study the effects of compressing images on data storage and 2D DIC measurements, and the effectiveness are compared between JPEG and JPEG 2000 image compression techniques. A cantilever-beam is adopted to experimentally investigate the impacts of the compressed image sets on 3D surface deformation and vibration response measurement. The proposed framework is a valuable option for DIC vibration measurement with high efficiency, the JPEG lossy compression and JPEG 2000 lossless compression each has its own characteristics in performance of DIC based vibration measurement.

Robust data hiding for JPEG images with invertible neural network

JPEG compression will cause severe distortion to the shared compressed image, which brings great challenges to extracting messages correctly from the stego image. To address such challenges, we propose a novel end-to-end robust data hiding scheme for JPEG images. The embedding and extracting secret messages on the quantized discrete cosine transform (DCT) coefficients are implemented by the bi-directional process of the invertible neural network (INN), which can provide intrinsic robustness against lossy JPEG compression. We design a JPEG compression attack module to simulate the JPEG compression process, which helps the network automatically learn how to recover the secret message from JPEG compressed image. Experimental results have demonstrated that our method achieves strong robustness against lossy JPEG compression, and also significantly improves the security compared with the existing data hiding methods on the premise of ensuring image quality and high capacity. For example, the detection error of our method against XuNet has been increased by 3.45% over the existing data hiding methods.

A Model-Driven Deep Unfolding Method for JPEG Artifacts Removal.

Deep learning-based methods have achieved notable progress in removing blocking artifacts caused by lossy JPEG compression on images. However, most deep learning-based methods handle this task by designing black-box network architectures to directly learn the relationships between the compressed images and their clean versions. These network architectures are always lack of sufficient interpretability, which limits their further improvements in deblocking performance. To address this issue, in this article, we propose a model-driven deep unfolding method for JPEG artifacts removal, with interpretable network structures. First, we build a maximum posterior (MAP) model for deblocking using convolutional dictionary learning and design an iterative optimization algorithm using proximal operators. Second, we unfold this iterative algorithm into a learnable deep network structure, where each module corresponds to a specific operation of the iterative algorithm. In this way, our network inherits the benefits of both the powerful model ability of data-driven deep learning method and the interpretability of traditional model-driven method. By training the proposed network in an end-to-end manner, all learnable modules can be automatically explored to well characterize the representations of both JPEG artifacts and image content. Experiments on synthetic and real-world datasets show that our method is able to generate competitive or even better deblocking results, compared with state-of-the-art methods both quantitatively and qualitatively.

Adversarial robust image steganography against lossy JPEG compression

Traditional image steganography assumes that the transmission channel is lossless, thus it can not assure the correct extraction of hidden secret messages in real social networks. Recently, some robust image steganographic methods have been proposed to resist lossy JPEG compression. In this paper, we first introduce adversarial mechanism into robust image steganography to enhance steganography security. Instead of modifying embedding costs in traditional adversarial embedding steganography, we propose adversarial dither modulation algorithm according to the cover gradients calculated from a pre-trained CNN (Convolutional Neural Networks)-based steganalyzer. This algorithm determines the dither directions via the signs and magnitudes of the cover gradients. In addition, we carefully select proper embedding elements according to the prior information of the transmission channel to obtain a good tradeoff between “robustness” and “security” for robust steganography. Experimental results show that the proposed method can achieve significant improvements on security while preserving very good robustness compared with related work.

A Target Corner Detection Algorithm Based on the Fusion of FAST and Harris

Corner detection is a common method to obtain image features, and the detection effect influences the performance of matching and tracking directly. A FAST-Harris fusion corner detection algorithm is proposed to improve the shortcomings of the Harris algorithm, such as the low detection accuracy and low positioning accuracy, and a corner detection fusion model is established. First, the detected target image is padded, and then the FAST algorithm is used with a 25% reduced contrast points to achieve fast capture roughly; in this way, a candidate corner set is obtained. Then, screening the candidate corner is set one by one by calculating the response function of the Harris with Scharr operator to achieve capture accurately. Finally, the real corners are obtained using SAD for nonmaximum suppression. The positioning error, error detection rate, robustness, and running time of corner detection are obtained by the PyCharm platform. Compared with Harris, the error detection rate and localization error of the algorithm are reduced by 16.89% and 42.04%, respectively. Compared with 8 popular corner detection algorithms, the error detection rate and localization error of the algorithm in this paper are the lowest, which are 24.60% and 1.42 pixels. The robust performance in lossy JPEG compression is the best, with 17.37% shorter running time than Harris algorithm. The method in this paper can be used in scenarios such as autonomous driving and image search services.

Blockchain based security framework for sharing digital images using reversible data hiding and encryption

Security is an important issue in current and next-generation networks. Blockchain will be an appropriate technology for securely sharing information in next-generation networks. Digital images are the prime medium attacked by cyber attackers. In this paper, a blockchain based security framework is proposed for sharing digital images in a multi user environment. The proposed framework uses reversible data hiding and encryption as component techniques. A novel high capacity reversible data hiding scheme is also proposed to protect digital images. Reversible data hiding in combination with encryption protects the confidentiality, integrity and authentication of digital images. In the proposed technique, the digital image is compressed first to create room for data hiding, then the user signature is embedded; afterwards the whole image is encrypted. For compression, JPEG lossy compression is used to create high capacity. For encryption, any symmetric block cipher or stream cipher can be used. Experimental results show that the proposed blockchain based framework provides high security and the proposed reversible data hiding scheme provides high capacity and image quality.

Edge-Aware Extended Star-Tetrix Transforms for CFA-Sampled Raw Camera Image Compression.

Codecs using spectral-spatial transforms efficiently compress raw camera images captured with a color filter array (CFA-sampled raw images) by changing their RGB color space into a decorrelated color space. This study describes two types of spectral-spatial transform, called extended Star-Tetrix transforms (XSTTs), and their edge-aware versions, called edge-aware XSTTs (EXSTTs), with no extra bits (side information) and little extra complexity. They are obtained by (i) extending the Star-Tetrix transform (STT), which is one of the latest spectral-spatial transforms, to a new version of our previously proposed wavelet-based spectral-spatial transform and a simpler version; (ii) considering that each 2D predict step of the wavelet transform is a combination of two 1D diagonal or horizontal-vertical transforms; (iii) weighting the transforms along the edge directions in the images. Compared with XSTTs, the EXSTTs can decorrelate CFA-sampled raw images well: they reduce the difference in energy between the two green components by about 3.38-30.08 % for high-quality camera images and 8.97-14.47 % for mobile phone images. The experiments on JPEG 2000-based lossless and lossy compression of CFA-sampled raw images show better performance than conventional methods. For high-quality camera images, the XSTTs/EXSTTs produce results equal to or better than the conventional methods: especially for images with many edges, the type-I EXSTT improves them by about 0.03-0.19 bpp in average lossless bitrate and the XSTTs improve them by about 0.16-0.96 dB in average Bjøntegaard delta peak signal-to-noise ratio. For mobile phone images, our previous work perform the best, whereas the XSTTs/EXSTTs show similar trends to the case of high-quality camera images.

Enhanced Binary MQ Arithmetic Coder with Look-Up Table

Binary MQ arithmetic coding is widely used as a basic entropy coder in multimedia coding system. MQ coder esteems high in compression efficiency to be used in JBIG2 and JPEG2000. The importance of arithmetic coding is increasing after it is adopted as a unique entropy coder in HEVC standard. In the binary MQ coder, arithmetic approximation without multiplication is used in the process of recursive subdivision of range interval. Because of the MPS/LPS exchange activity that happens in the MQ coder, the output byte tends to increase. This paper proposes an enhanced binary MQ arithmetic coder to make use of look-up table (LUT) for (A × Qe) using quantization skill to improve the coding efficiency. Multi-level quantization using 2-level, 4-level and 8-level look-up tables is proposed in this paper. Experimental results applying to binary documents show about 3% improvement for basic context-free binary arithmetic coding. In the case of JBIG2 bi-level image compression standard, compression efficiency improved about 0.9%. In addition, in the case of lossless JPEG2000 compression, compressed byte decreases 1.5% using 8-level LUT. For the lossy JPEG2000 coding, this figure is a little lower, about 0.3% improvement of PSNR at the same rate.

Cellular automata-based digital image scrambling under JPEG compression attack

Communication and storage technologies have greatly advanced in recent years, providing ever increased bandwidths and storage capacities respectively. But, efficient utilization of resources like network bandwidth and data storage is relevant even today. More and more data are generated every unit of time and this enormous data is sent over networks and stored in local and/or cloud storage devices. Compression techniques are used to compress or reduce the size of data so as to make efficient use of these resources. Compression techniques are basically of two types; lossless compression used by applications that require integrity of the data is preserved and lossy compression used by applications in which subtle changes in data either go unnoticed or does not affect the semantics of information. In this paper, we analyze the effect of JPEG lossy compression on scrambled images and propose a cellular automata-based method to recover the quality of de-scrambled images.

Автоматизоване розпізнавання міської рослинності та водойм за знімками супутника «Jilin-1А»

The results of testing the developed techniques for automatized recognition of vegetation and water bodies on the urban territory by multispectral images from the Jilin-1А satellite are given.The research included automatized recognition of vegetation and water bodies on the selected observation territory based on images with super high spatial resolution in the visual and infrared range and consequent comparison of the obtained results with the results of visual decoding. The obtained results of processing the images from the Jilin-1А satellite in accordance with the proposed techniques confirmed the sufficiently high accuracy of automatized edge enhancement of recognized objects as compared to the results of interactive visual recognition of these images. Different test areas provided a good separation of vegetation and water types with the same thresholding customization. The accuracy of automatized classification of vegetation and water bodies (without considering the standard errors) for different test areas was within 81...92%, and values of kappa-coefficient were within 0.68 to 0.85. Comparison of normalized index images received from Jilin-1А and Sentinel-2A satellites showed slight discordance in NDVI values and significant discordances for NDWI and MNDWI that are caused by the usage of different spectral channels (SWIR and NIR). These discordances can be sufficiently reduced when using correction coefficients.Analysis of the influence of output image resolution reduction (from 10 to 8 bit) and subsequent informational compressing (JPEG lossy and JPEG2000 lossless) on results of automatized recognition of vegetation and water bodies confirmed the validity and efficiency of these techniques. The volume of saved and transmitted files significantly decreased (in 80…100 times) with a slight reduction of classification accuracy (by 1...2 %). The proposed techniques make it possible to increase significantly the efficiency and probability of renewing maps of big cities and to reduce financial expenditures as compared to the traditional ground GPS-surveying and aerosurveying. The high-level automatization of image processing and minimization of necessary calculations (as compared to techniques that use complex classifiers and neural networks) allow to implement the developed technique as a geographic information web service that satisfies the needs of a wide circle of government services and commercial structures and can be useful for megalopolis population and tourists.

Second compression for pixelated images under edge-based compression algorithms: JPEG-LS as an example

This paper details the examination of a particular case of data compression, where the compression algorithm removes the redundancy from data, which occurs when edge-based compression algorithms compress (previously compressed) pixelated images. The newly created redundancy can be removed using another round of compression. This work utilized the JPEG-LS as an example of an edge-based compression algorithm for compressing pixelated images. The output of this process was subjected to another round of compression using a more robust but slower compressor (PAQ8f). The compression ratio of the second compression was, on average, 18%, which is high for random data. The results of the second compression were superior to the lossy JPEG. Under the used data set, lossy JPEG needs to sacrifice 10% on average to realize nearly total lossless compression ratios of the two-successive compressions. To generalize the results, fast general-purpose compression algorithms (7z, bz2, and Gzip) were used too.

A Novel Multilevel Lossy Compression Algorithm for Grayscale Images Inspired by the Synthesization of Biological Protein Sequences

Enormous number of images are generated daily in all areas of life, including social media, medical and navigation images. Moreover, the development of smart phones among other specialized media-capturing devices has witnessed great advances during the last decade. Consequently, the storage, transmission, and analysis of images become essential and frequent tasks. Thus, various research efforts tried to address the image compression problem from different computational perspectives. This article presents a novel multilevel lossy compression algorithm for grayscale images, namely Image-as-Protein ( IaP ), that is inspired by the translation of DNA sequences into protein sequences that occurs inside live beings. Because of the high similarity of the resulting textual protein sequence, it can be tackled by general text compression techniques with competitive compression ratios. Various qualitative comparisons and quantitative measures such as BPP , SSIM and PSNR have been carried out on multiple grayscale image benchmark datasets. The experimental results showed that the proposed algorithm is promising compared to the famous JPEG lossy image compression standard.

The DICOM CT Image Compression Based On Enhanced Lossless Prediction And Multilevel Thresholding Based Hybrid Cuckoo Search With Hill Climbing (CS-HC) Algorithm Based Segmentation

In computer vision applications, image segmentation is a common image processing step. It is used to separate pixels into different groups. The rise in the threshold count would hinder the segmentation phase of images. At the same time, in the field of threshold implementation in the image, it becomes an NT concern. This thesis suggests a multilevel threshold based on optimization techniques to remove ROI and uses enhanced lossless prediction algorithm to compress DICOM images in telemedicine applications. The hybrid Cuckoo search with hill climbing (CS-HC) algorithm strengthens the process used by the search agent to update the optimal solution. This algorithm calculates the threshold value. The superior results are produced by the proposed multilevel level thresholding based on CS-HC, as seen by the simulation results. Optimization is efficient and it has a high degree of convergence. Effective results are provided by the proposed lossless compression algorithm based on classification and blending estimation as compared with JPEG lossless and lossy compression techniques. With various threshold values, the algorithm 's efficiency is checked. To apply this algorithm, Matlab2010a is used and DICOM photos are used to validate it.

Deep dilated and densely connected parallel convolutional groups for compression artifacts reduction

A widely used JPEG lossy compression standard introduces different artifacts, particularly, blocking, ringing and blurring effects in the compressed image. Existing algorithms focus on deblocking and discard the ringing, blurring artifacts that produce a visually unpleasant images. These artifacts must be eliminated to restore visually pleasing output while decompressing the image. In this paper, we propose a novel 10-layer network with parallel convolutional groups based on deep dilated convolutions and dense connections (D3C Net) to reduce blocking, ringing and blurring artifacts arise due to JPEG compression. The dilated convolution helps to expand receptive field exponentially without increasing the complexity of the network, thereby relieving the network from the training burden. Moreover parallel convolutional layers are integrated into network for efficient feature representation and model parallelism. In addition, the skip connections allow the information to be back-propagated efficiently to bottom layers and tackles the problem of vanishing gradients. Furthermore, the gradient explosion problem is tackled by an end-to-end mapping between the JPEG decompressed image and the corresponding residual image. Different from the existing algorithms for compression artifacts removal which usually train for a specific compression factor, our D3C Net is able to handle compression artifacts with unknown compression factors (i.e., blind JPEG compression artifacts removal). The subjective and objective performance analysis with metrics PSNR, SSIM and PSNR-B demonstrates that the proposed network produces significant performance improvement when compared to state-of-the-art methods.

JSNet: A simulation network of JPEG lossy compression and restoration for robust image watermarking against JPEG attack

Deep learning-based watermarking methods have achieved a better performance in capacity and invisibility than some traditional methods. However, their robustness against JPEG lossy compression attack is still to be improved. To enhance the robustness and construct an end-to-end method, it is urgent to simulate the JPEG lossy compression by a neural network and then introduce it into the deep learning-based watermarking methods. In this paper, a JPEG simulation network JSNet is proposed to reappear the whole procedure of the JPEG lossy compression and restoration except entropy encoding as realistically as possible. The steps of sampling, DCT, and quantization are modeled by the max-pooling layer, convolution layer, and 3D noise-mask, respectively. The proposed JSNet can simulate JPEG lossy compression with any quality factors. To verify the proposed JSNet in improving the robustness against JPEG compression attack, a CNN-based robust watermarking network (CRWNet) is proposed as an application example. The end-to-end CRWNet contains three subnetworks, i.e., embedding subnetwork, JSNet, and extraction subnetwork. Here. the JSNet is regarded as an attack module in the pipeline. Experimental results on two publicly available datasets (ImageNet, BossBase) demonstrate that: (a) the proposed JSNet can well simulate JPEG lossy compression under any quality factors with small Root mean square error (RMSE) values; (b) the proposed CRWNet considering JSNet has achieved an average 30.6 percent advantage over the basic model without consideration of JSNet.