Median filtering detection has recently drawn much attention in image editing and image anti-forensic techniques. Current image median filtering forensics algorithms mainly extract features manually. To deal with the challenge of detecting median filtering from small-size and compressed image blocks, by taking into account of the properties of median filtering, we propose a median filtering detection method based on convolutional neural networks (CNNs), which can automatically learn and obtain features directly from the image. To our best knowledge, this is the first work of applying CNNs in median filtering image forensics. Unlike conventional CNN models, the first layer of our CNN framework is a filter layer that accepts an image as the input and outputs its median filtering residual (MFR). Then, via alternating convolutional layers and pooling layers to learn hierarchical representations, we obtain multiple features for further classification. We test the proposed method on several experiments. The results show that the proposed method achieves significant performance improvements, especially in the cut-and-paste forgery detection.Index Terms-Convolutional neural networks, deep learning, hierarchical representations, median filtering forensics.
Background Growing evidences indicate that circular RNAs (circRNAs) play an important role in the regulation of biological behavior of tumor. We aim to explore the role of circRNA in glioma and elucidate how circRNA acts. Methods Real-time PCR was used to examine the expression of circPTN in glioma tissues and normal brain tissues (NBT). Assays of dual- luciferase reporter system, biotin label RNA pull-down and FISH were used to determine that circPTN could sponge miR-145-5p and miR-330-5p. Tumor sphere formation assay was performed to determine self- renewal of glioma stem cell (GSCs). Cell counting Kit-8 (CCK8), EdU assay and flow cytometry were used to investigate proliferation and cell cycle. Intracranial xenograft was established to determine how circPTN impacts in vivo. Tumor sphere formation assay was performed to determine self- renewal of glioma stem cell (GSCs). Results We demonstrated circPTN was significantly higher expression in glioma tissues and glioma cell lines, compared with NBT and HEB (human astrocyte). In gain- and loss-of-function experiments, circPTN significantly promoted glioma growth in vitro and in vivo. Furthermore, we performed dual-luciferase reporter assays and RNA pull-down assays to verify that circPTN acts through sponging miR-145-5p and miR-330-5p. Increasing expression of circPTN rescued the inhibition of proliferation and downregulation of SOX9/ITGA5 in glioma cells by miR-145-5p/miR-330-5p. In addition, we found that circPTN promoted self-renewal and increased the expression of stemness markers (Nestin, CD133, SOX9, and SOX2) via sponging miR-145-5p. Moreover, this regulation was disappeared when circPTN binding sites in miR-145-5p were mutated. Conclusions Our results suggest that circPTN is an oncogenic factor that acts by sponging miR-145-5p/miR-330-5p in glioma.
We propose a large-margin Gaussian Mixture (L-GM) loss for deep neural networks in classification tasks. Different from the softmax cross-entropy loss, our proposal is established on the assumption that the deep features of the training set follow a Gaussian Mixture distribution. By involving a classification margin and a likelihood regularization, the L-GM loss facilitates both a high classification performance and an accurate modeling of the training feature distribution. As such, the L-GM loss is superior to the softmax loss and its major variants in the sense that besides classification, it can be readily used to distinguish abnormal inputs, such as the adversarial examples, based on their features' likelihood to the training feature distribution. Extensive experiments on various recognition benchmarks like MNIST, CIFAR, ImageNet and LFW, as well as on adversarial examples demonstrate the effectiveness of our proposal.
In this paper, we present a superpixel segmentation algorithm called linear spectral clustering (LSC), which is capable of producing superpixels with both high boundary adherence and visual compactness for natural images with low computational costs. In LSC, a normalized cuts-based formulation of image segmentation is adopted using a distance metric that measures both the color similarity and the space proximity between image pixels. However, rather than directly using the traditional eigen-based algorithm, we approximate the similarity metric through a deliberately designed kernel function such that pixel values can be explicitly mapped to a high-dimensional feature space. We then apply the conclusion that by appropriately weighting each point in this feature space, the objective functions of the weighted K-means and the normalized cuts share the same optimum points. Consequently, it is possible to optimize the cost function of the normalized cuts by iteratively applying simple K-means clustering in the proposed feature space. LSC possesses linear computational complexity and high memory efficiency, since it avoids both the decomposition of the affinity matrix and the generation of the large kernel matrix. By utilizing the underlying mathematical equivalence between the two types of seemingly different methods, LSC successfully preserves global image structures through efficient local operations. Experimental results show that LSC performs as well as or even better than the state-of-the-art superpixel segmentation algorithms in terms of several commonly used evaluation metrics in image segmentation. The applicability of LSC is further demonstrated in two related computer vision tasks.
Novel NiO@ZnO heterostructured nanotubes (NTs) were fabricated by the coelectrospinning method, consisting of external hexagonal ZnO shell and internal cubic NiO NTs. They are carefully investigated by scanning electron microscopy, transmission electron microscopy, scanning transmission electron microscopy, energy-dispersive X-ray spectroscopy mapping, X-ray diffraction, and X-ray photoelectron spectroscopy techniques. A reasonable formation mechanism of the hierarchical NiO@ZnO NTs is proposed, which is discussed from the view of degradation temperature of different polymers and the amount of inorganic salts. They were then explored for fabrication of H(2)S gas sensors. The gas sensing test reveals that compared with the pure ZnO, NiO, and the ZnO-NiO mixed gas sensors, hierarchical gas sensor exhibits highly improved sensing performances to dilute hydrogen sulfide (H(2)S) gas. The response of the optimum NiO@ZnO NTs sensor to 50 ppm H(2)S increases as high as 2.7-23.7 times compared to the other sensors, whereas the response and recovery times also become shorter considerably. These enhanced gas sensing properties are closely related to the change of nanostructure and activity of ZnO and NiO nanocrystals as well as combination of homo- and heterointerfaces in the optimum gas sensor, which are confirmed by a series of well-designed experiments.
Face image quality is an important factor affecting the accuracy of automatic face recognition. It is usually possible for practical recognition systems to capture multiple face images from each subject. Selecting face images with high quality for recognition is a promising stratagem for improving the system performance. We propose a learning to rank based framework for assessing the face image quality. The proposed method is simple and can adapt to different recognition methods. Experimental result demonstrates its effectiveness in improving the robustness of face detection and recognition.
Summary Appropriate regulation of crop seed germination is of significance for agriculture production. In this study, we show that TaJAZ1, most closely related to Arabidopsis JAZ3, negatively modulates abscisic acid (ABA)‐inhibited seed germination and ABA‐responsive gene expression in bread wheat. Biochemical interaction assays demonstrate that the C‐terminal part containing the Jas domain of TaJAZ1 physically interacts with TaABI5. Similarly, Arabidopsis jasmonate‐ZIM domain (JAZ) proteins also negatively modulate ABA responses. Further we find that a subset of JAZ proteins could interact with ABI5 using the luciferase complementation imaging assays. Choosing JAZ3 as a representative, we demonstrate that JAZ3 interacts with ABI5 in vivo and represses the transcriptional activation activity of ABI5. ABA application could abolish the enrichment of JAZ proteins in the ABA‐responsive gene promoter. Furthermore, we find that ABA application could induce the expression of jasmonate (JA) biosynthetic genes and then increase the JA concentrations partially dependent on the function of ABI5, consequently leading to the degradation of JAZ proteins. This study sheds new light on the crosstalk between JA and ABA in modulating seed germination in bread wheat and Arabidopsis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.