Search

Image recoloring is an emerging editing technique that can change the color style of an image by modifying pixel values without altering the original image content. With the rapid proliferation of social network and image editing techniques, recolored images (RIs) have raised new security issues in society. Existing detection methods have good performance in detecting RIs for certain categories of recoloring techniques. However, the performance on the handcrafted recoloring scenario is still poor due to the influence of human prior knowledge. To deal with this problem, we explore a solution from the perspective of chrominance texture artifacts to improve the generalization ability. The results of the analysis show that natural images (NIs) and RIs have textural disparities in different color components, especially in the chrominance components (i.e., Cb, Cr, and H). Based on such new prior knowledge of statistical discriminability, we propose a feature set to capture texture features in chrominance components for identifying RIs. Extensive experimental results show that the proposed method can accurately identify RIs with certain categories of recoloring techniques, and outperforms existing methods in the scenario of handcrafted recoloring.

Photo-response non-uniformity (PRNU), as a class of device fingerprint, plays a key role in the forgery detection/localization for visual media. The state-of-the-art PRNU-based forensics methods generally rely on the multi-scale trace analysis and result fusion, with Markov random field model. However, such hand-crafted strategies are difficult to provide satisfactory multi-scale decision, exhibiting a high false-positive rate. Motivated by this, we propose an end-to-end multi-scale decision fusion strategy, where a mapping from multi-scale forgery probabilities to binary decision is achieved by a supervised deep fully connected neural network. As the first time, the deep learning technology is employed in PRNU-based forensics for more flexible and reliable integration of multi-scale information. The benchmark experiments exhibit the state-of-the-art accuracy performance of our method in both pixel-level and image-level, especially for false positives. Additional robustness experiments also demonstrate the benefits of the proposed method in resisting noise and compression attacks.

The training and creation of deep learning model is usually costly, thus it can be regarded as an intellectual property (IP) of the model creator. However, malicious users who obtain high-performance models may illegally copy, redistribute, or abuse the models without permission. To deal with such security threats, a few deep neural networks (DNN) IP protection methods have been proposed in recent years. This paper attempts to provide a review of the existing DNN IP protection works and also an outlook. First, we propose the first taxonomy for DNN IP protection methods in terms of six attributes: scenario, mechanism, capacity, type, function, and target models. Then, we present a survey on existing DNN IP protection works in terms of the above six attributes, especially focusing on the challenges these methods face, whether these methods can provide proactive protection, and their resistances to different levels of attacks. After that, we analyze the potential attacks on DNN IP protection methods from the aspects of model modifications, evasion attacks, and active attacks. Besides, a systematic evaluation method for DNN IP protection methods with respect to basic functional metrics, attack-resistance metrics, and customized metrics for different application scenarios is given. Lastly, future research opportunities and challenges on DNN IP protection are presented.

The form of inorganic N in soil is governed by N transformations, but the relationship between the soil N transformation dynamics and the N uptake by plants is unclear. The effect of liming and nitrification inhibitor on crop N uptake efficiency (NUE) and N loss due to altered soil N processes was investigated using a 15 N tracing approach. Crops with different N uptake strategies (sweet potato (Dioscorea esculenta L.); an ammonium-N preferring plant and pakchoi (Brassica chinensis L.); a nitrate–N preferring plant) were grown in two agricultural soils with different pH values with or without lime and nitrification inhibitor (3,4-dimethylpyrazole phosphate, DMPP). We monitored the effects of 15 N-nitrate or 15 N-ammonium on the plant biomass and NUE, residual soil N, leachate, and emitted gas. The NUE of pakchoi in acidic soil was greater with nitrate than with ammonium; however, when lime was applied, the NUE of pakchoi increased significantly compared to ammonium alone. The NUE of pakchoi was not affected by N source in the alkaline soil, but application of DMPP together with ammonium significantly reduced NUE of pakchoi compared to ammonium alone. The N source had no effect on the NUE of sweet potato in acidic or alkaline soils, while NUE increased significantly after application of lime and ammonium (31.3%) and with application of DMPP (32.4%) compared to ammonium alone (30% in acidic soil and 23% in alkaline soil). Pakchoi NUE was significantly (p

Cost-friendly differential privacy (CDP) of smart meters can be preserved by an appropriate charging and discharging mechanism that uses rechargeable batteries (RBs) to generate Laplace distributed random noise. However, the existing CDP methods have several issues. Firstly, the maximum discharge rate of an RB requires to vary with the maximal consumption of houses. Secondly, the probability of an RB to charge/discharge depends on the demand, regardless of the state of charge (SoC) of an RB. Thirdly, in extreme SoC (near-empty or almost fully charged) of an RB, no noise added to the demand. To overcome these, we propose a mechanism in which a novel probability density function is designed to generate near Laplace distributed random noise. We also utilize a renewable energy source with small storage in cascade with an RB to enhance performance. Both theoretical analysis and simulations are performed to demonstrate the effectiveness of our proposed method.

The maturation of wireless connectivity, blockchain (distributed ledger technologies), and intelligent systems has fostered a comprehensive ecosystem for the Internet of Things (IoT). However, the growing volume of data generated by IoT devices creates substantial pressure on blockchain storage and computation capabilities, impeding the further development of the IoT ecosystem. Decentralizing data storage across multiple chains and utilizing cross-chain technology for data exchange eliminates the need for expensive centralized infrastructure, lowers data transfer costs, and improves accessibility. Hence, the issue of computational and storage pressure in blockchain can be improved. Nonetheless, the data of IoT devices are constantly updating, and ensuring consistency for dynamic data across heterogeneous chains remains a significant challenge. To address the aforementioned challenge, we propose a blockchain-based distributed and lightweight data consistency verification model (BDCA), which leverages a batch verification dynamic Merkle hash tree (BV-MHT) and an advanced gamma multi-signature scheme (AGMS) to enable consistent verification of dynamic data while ensuring secure and private data transmission. The AGMS scheme is reliable and robust based on security analysis while the dependability and consistency of BDCA are verified through inductive reasoning. Experimental results indicate that BDCA outperforms CPVPA and Fortress in communication and computation overhead for data preprocessing and auditing in a similar condition, and the AGMS scheme exhibits superior performance when compared to other widely adopted multi-signature schemes such as Cosi, BLS, and RSA. Furthermore, BDCA provides up to 99% data consistency guarantees, demonstrating its practicality.

Soil moisture content (SWC) can change the diversity and composition of soil fungal communities by affecting soil texture and soil nutrients. To explore the response of soil fungal communities to moisture in the grassland ecosystem on the south shore of Hulun Lake, we set up a natural moisture gradient that was subdivided into high (HW), medium (MW), and low (LW) water contents. Vegetation was investigated by quadrat method, and aboveground biomass was collected by the mowing method. Soil physicochemical properties were obtained by internal experiments. The composition of the soil fungal community was determined using high-throughput sequencing technology. The results showed significant differences in soil texture, nutrients, and fungal species diversity under the moisture gradients. Although there was significant clustering of fungal communities in different treatments, the fungal community composition was not significantly different. According to the phylogenetic tree, the Ascomycota and Basidiomycota were the most important branches. The fungal species diversity was smaller when SWC was higher, and in this environment (HW), the fungal-dominant species were significantly related to SWC and soil nutrients. At this time, soil clay formed a protective barrier for the survival of the dominant classes Sordariomycetes and Dothideomycetes and increased their relative abundance. In summary, the fungal community responded significantly to SWC on the southern shore of the Hulun Lake ecosystem in Inner Mongolia, China, and the fungal community composition of the HW group was stable and easier to survive.

The effective recognition of patterns from blurred images presents a fundamental difficulty for many practical vision tasks. In the era of deep learning, the main ideas to cope with this difficulty are data augmentation and deblurring. However, both facing issues such as inefficiency, instability, and lack of explainability. In this paper, we explore a simple but effective way to define invariants from blurred images, without data augmentation and deblurring. Here, the invariants are designed from Fractional Moments under Projection operators (FMP), where the blur invariance and rotation invariance are guaranteed by the general theorem of blur invariants and the Fourier-domain rotation equivariance, respectively. In general, the proposed FMP not only bears a simpler explicit definition, but also has useful representation properties including orthogonality, statistical flexibility, as well as the combined invariance of blurring and rotation. Simulation experiments are provided to demonstrate such properties of our FMP, revealing the potential for small-scale robust vision problems.

When visual sensor networks (VSNs) enter daily life in society, they not only bring great convenience but also cause people to worry about privacy. Traditional image encryption often displays the effect of the enter snowflake in encrypted images playing a part in privacy protection, but the usability of directly browsing images taken by nodes in VSNs is sacrificed. Recently, thumbnail-preserving encryption (TPE) has been proposed to balance image usability and privacy. However, the Markov chain’s limited connectivity and high time cost, which result in low efficiency, may preclude its application in VSNs. Motivated by this, we propose a novel TPE scheme to protect the image collected in VSNs privacy efficiently. We first deeply study the bijective relationship between the three-pixel group and its rank, and then propose a new number method for both, based on which we design the efficient rank mapping. Subsequently, an efficient thumbnail-preserving image encryption via three-pixel rank mapping (E-TPE) is designed, achieving NR security and enhancing the Markov chain’s connectivity. The experiments demonstrate the scheme’s efficiency, that is, it is the currently quickest ideal-TPE scheme with NR security, and that a good balance is achieved between usability and privacy.

In recent years, many backdoor attacks based on training data poisoning have been proposed. However, in practice, those backdoor attacks are vulnerable to image compressions. When backdoor instances are compressed, the feature of specific backdoor trigger will be destroyed, which could result in the backdoor attack performance deteriorating. In this paper, we propose a compression-resistant backdoor attack based on feature consistency training. To the best of our knowledge, this is the first backdoor attack that is robust to image compressions. First, both backdoor images and their compressed versions are input into the deep neural network (DNN) for training. Then, the feature of each image is extracted by internal layers of the DNN. Next, the feature difference between backdoor images and their compressed versions are minimized. As a result, the DNN treats the feature of compressed images as the feature of backdoor images in feature space. After training, the backdoor attack against DNN is robust to image compression. Furthermore, we consider three different image compressions (i.e., JPEG, JPEG2000, WEBP) in feature consistency training, so that the backdoor attack is robust to multiple image compression algorithms. Experimental results demonstrate the effectiveness and robustness of the proposed backdoor attack. When the backdoor instances are compressed, the attack success rate of common backdoor attack is lower than 10%, while the attack success rate of our compression-resistant backdoor is greater than 97%. The compression-resistant attack is still robust even when the backdoor images are compressed with low compression quality. In addition, extensive experiments have demonstrated that, our compression-resistant backdoor attack has the generalization ability to resist image compression which is not used in the training process.

Non-Fungible Tokens (NFTs) do hold the promise of providing Web3 with the opportunity for self-sovereignty of users' physical assets. However, existing NFT marketplaces lack a generic design that allows the value of assets to flow efficiently. In this paper, we propose a generic NFT architecture for Web3. The architecture supports the rapid development of the upper application environment and automated value mapping of the underlying physical asset environment. To connect these two environments, a generic connecter has been designed to provide flexible storage for mapping data management, and to support universal cross-chain transactions. With these features, the values of heterogeneous physical assets can coexist in a unified Web3 world, and rich value transfer services can be developed on demand. This paper discusses the background of the proposed architecture, the open problems and our initial solution, as well as our design principles and advantages, and finally validates this novel NFT architecture.

BackgroundThe patients with chronic low back pain (CLBP) showed impaired postural control, especially in challenging postural task. The dorsolateral prefrontal cortex (DLPFC) is reported to involve in the complex balance task, which required considerable attentional control. The effect of intermittent theta burst stimulation (iTBS) over the DLPFC to the capacity of postural control of CLBP patients is still unknown.MethodsParticipants diagnosed with CLBP received a single-session iTBS over the left DLPFC. All the participants completed the postural control tasks of single-leg (left/right) standing before and after iTBS. The activation changes of the DLPFC and M1 before and after iTBS were recorded by functional near-infrared spectroscopy (fNIRS). The activation pattern of the trunk [transversus abdominis (TrA), superficial lumbar multifidus (SLM)] and leg [tibialis anterior (TA), gastrocnemius medialis (GM)] muscles including root mean square (RMS) and co-contraction index (CCI) during single-leg standing were measured by surface electromyography (sEMG) before and after the intervention. The paired t-test was used to test the difference before and after iTBS. Pearson correlation analyses were performed to test the relationship between the oxyhemoglobin concentration and sEMG outcome variables (RMS and CCI).ResultsOverall, 20 participants were recruited. In the right-leg standing condition, compared with before iTBS, the CCI of the right TrA/SLM was significantly decreased (t = −2.172, p = 0.043), and the RMS of the right GM was significantly increased (t = 4.024, p = 0.001) after iTBS. The activation of the left DLPFC (t = 2.783, p = 0.012) and left M1 (t = 2.752, p = 0.013) were significantly decreased and the relationship between the left DLPFC and M1 was significant after iTBS (r = 0.575, p = 0.014). Correlation analysis showed the hemoglobin concentration of M1 was negatively correlated with the RMS of the right GM (r = −0.659, p = 0.03) and positively correlated between CCI of the right TrA/SLM (r = 0.503, p = 0.047) after iTBS. There was no significant difference in the brain or muscle activation change in the left leg-standing condition between before and after iTBS.ConclusionIntermittent theta burst stimulation over the left DLPFC seems to be able to improve the muscle activation pattern during postural control ability in challenging postural task, which would provide a new approach to the treatment of CLBP.

Introduction: This study aimed to assess the influence of experimental warming and fertilization on rice yield and paddy methane emissions. Methods: A free-air temperature increase system was used for the experimental warming treatment (ET), while the control treatment used ambient temperature (AC). Each treatment contained two fertilization strategies, (i) normal fertilization with N, P and K fertilizers (CN) and (ii) without N fertilizer input (CK). Results: The yield was remarkably dictated by fertilization (p < 0.01), but not warming. Its value with CN treatment increased by 76.24% compared to CK. Also, the interactive effect of warming and fertilization on CH4 emissions was insignificant. The seasonal emissions from warming increased by 36.93% compared to AC, while the values under CN treatment increased by 79.92% compared to CK. Accordingly, the ET-CN treatment obtained the highest CH4 emissions (178.08 kg ha−1), notably higher than the other treatments. Also, the results showed that soil fertility is the main driver affecting CH4 emissions rather than soil microorganisms. Conclusions: Fertilization aggravates the increasing effect of warming on paddy methane emissions. It is a daunting task to optimize fertilization to ensure yield and reduce methane emissions amid global warming.

The location information for tasks may expose sensitive information, which impedes the practical use of mobile crowdsensing (MCS) in the industrial Internet. In this paper, to our knowledge, we are the first to discuss the privacy protection of task locations and propose a codebook-based task allocation mechanism to protect it. Considering the cost of system utility caused by privacy protection technology, the tradeoff between local privacy and system utility is formalized as a multi-objective optimization problem. The optimal solution is theoretically derived, and the optimal task allocation scheme is obtained. In addition, the selected allocation codebook (SAC) method is introduced to solve the problem of high computational resource consumption in the task allocation process and protect the task location privacy to some extent. The experimental results show that the SAC method sacrifices system utility but improves the privacy protection for task locations by 60% on average.

With the popularity of cloud storage services, people are increasingly accustomed to storing images in the cloud. However, cloud storage services raise privacy concerns, e.g., leakage of images to unauthorized third parties and service providers may exploit image detection technologies to portrait users without permission. Although privacy concerns can be solved by encrypting images before they are uploaded to the cloud, traditional encryption methods significantly affect the usability and user experience, for example, users cannot preview images in the cloud. Recently, Marohn et al. proposed two approximate thumbnail-preserving encryption schemes, called DRPE and TPE-LSB, to balance the privacy and usability of images in the cloud. However, both schemes have defects that either the decryption may fail or the ciphertext images have poor performance in perceived quality and too many noise points after decryption. To this end, we pertinently propose a high-fidelity thumbnail-preserving encryption scheme (HF-TPE). Compared with the previous works, on the one hand, the HF-TPE scheme not only ensures the correct decryption of ciphertext images, but also makes the ciphertext thumbnails more close to the plaintext images perceptually. On the other hand, the decrypted thumbnails have lower noise intensity and upper limit of the number of noises. In addition, simulation experiments further show that the HF-TPE scheme can guarantee users’ usability.