Qingxiao Guan

ListNet-based object proposals ranking

Abstract In object detection, object proposal methods have been widely used to generate candidate regions which may contain objects. Object proposal based on superpixel merging is one kind of object proposal methods, and the merging strategies of superpixels have been extensively explored. However, the ranking of generated candidate proposals still remains to be further studied. In this paper, we formulate the ranking of object proposals as a learning to rank problem, and propose a novel object proposals ranking method based on ListNet. In the proposed method, Selective Search, which is one of the state-of-the-art object proposal methods based on superpixel merging, is adopted to generate the candidate proposals. During the superpixel merging process, five discriminative objectness features are extracted from superpixel sets and the corresponding bounding boxes. Then, to weight each feature, a linear neural network is learned based on ListNet. Consequently, objectness scores can be computed for final candidate proposals ranking. Extensive experiments demonstrate the effectiveness and robustness of the proposed method.

Constructing Near-optimal Double-layered Syndrome-Trellis Codes for Spatial Steganography

In this paper, we present a new kind of near-optimal double-layered syndrome-trellis codes (STCs) for spatial domain steganography. The STCs can hide longer message or improve the security with the same-length message comparing to the previous double-layered STCs. In our scheme, according to the theoretical deduction we can more precisely divide the secret payload into two parts which will be embedded in the first layer and the second layer of the cover respectively with binary STCs. When embed the message, we encourage to realize the double-layered embedding by ±1 modifications. But in order to further decrease the modifications and improve the time efficient, we allow few pixels to be modified by ±2. Experiment results demonstrate that while applying this double-layered STCs to the adaptive steganographic algorithms, the embedding modifications become more concentrative and the number decreases, consequently the security of steganography is improved.

Copy-move forgery detection based on convolutional kernel network

Conventional copy-move forgery detection methods mostly make use of hand-crafted features to conduct feature extraction and patch matching. However, the discriminative capability and the invariance to particular transformations of hand-crafted features are not good enough, which imposes restrictions on the performance of copy-move forgery detection. To solve this problem, we propose to utilize Convolutional Kernel Network to conduct copy-move forgery detection. Convolutional Kernel Network is a kind of data-driven local descriptor with the deep convolutional architecture. It can achieve competitive performance for its excellent discriminative capability. To well adapt to the condition of copy-move forgery detection, three significant improvements are made: First of all, our Convolutional Kernel Network is reconstructed for GPU. The GPU-based reconstruction results in high efficiency and makes it possible to apply to thousands of patches matching in copy-move forgery detection. Second, a segmentation-based keypoint distribution strategy is proposed to generate homogeneous distributed keypoints. Last but not least, an adaptive oversegmentation method is adopted. Experiments on the publicly available datasets are conducted to testify the state-of-the-art performance of the proposed method.

Highly accurate real-time image steganalysis based on GPU

With the development of steganography, it is required to build high-dimensional feature spaces to detect those sophisticated steganographic schemes. However, the huge time cost prevents the practical deployment of high-dimensional features for steganalysis. SRM and DCTR are important steganalysis feature sets in spatial domain and JPEG domain, respectively. It is necessary to accelerate the extraction of DCTR and SRM to make them more usable in practice, especially for some real-time applications. In this paper, both DCTR and SRM are implemented on the GPU device to exploit the parallel power of the GPU and some optimization methods are presented. For implementation of DCTR, we first utilize the separability and symmetry of two-dimensional discrete cosine transform in decompression and convolution. Then, in order to make phase-aware histograms favorable for parallel GPU processing, we convert them into ordinary 256-dimensional histograms. For SRM, in computing residuals, we specify the computation sequence and spilt the inseparable two-dimensional kernel into several row vectors. When computing the four-dimensional co-occurrences, we convert them into one-dimensional histograms which are more suitable for parallel computing. The experimental results show that the proposed methods can greatly accelerate the extraction of DCTR and SRM, especially for images of large size. Our methods can be applied to the real-time steganalysis system.

Improving GFR Steganalysis Features by Using Gabor Symmetry and Weighted Histograms

The GFR (Gabor Filter Residual) features, built as histograms of quantized residuals obtained with 2D Gabor filters, can achieve competitive detection performance against adaptive JPEG steganography. In this paper, an improved version of the GFR is proposed. First, a novel histogram merging method is proposed according to the symmetries between different Gabor filters, thus making the features more compact and robust. Second, a new weighted histogram method is proposed by considering the position of the residual value in a quantization interval, making the features more sensitive to the slight changes in residual values. The experiments are given to demonstrate the effectiveness of our proposed methods.

Embedding Strategy for Batch Adaptive Steganography

In this paper, we present a new embedding strategy for batch adaptive steganography. This strategy can make up for the problem when applying batch steganography to adaptive steganography and determine the sub-batch of cover images to carry the total message. Firstly, we define a secure factor \(\alpha \) to evaluate the embedding security. Then, we utilize the secure factor \(\alpha \) to calculate the corresponding payloads for each image based on distortion-limited sender (maximizing the payload while introducing a fixed total distortion) and fit the relation curve for the secure factor and the corresponding payload. When embedding, the images with large size and more upward convex relation curve are employed to carry the total message. Experimental results show that fitting curves vary according to images and employing images with more upward convex relation curve to carry the total message can improve the secure performance effectively.

Image Forgery Localization based on Multi-Scale Convolutional Neural Networks

In this paper, we propose to utilize Convolutional Neural Networks (CNNs) and the segmentation-based multi-scale analysis to locate tampered areas in digital images. First, to deal with color input sliding windows of different scales, we adopt a unified CNN architecture. Then, we elaborately design the training procedures of CNNs on sampled training patches. With a set of tampering detectors based on CNNs for different scales, a series of complementary tampering possibility maps can be generated. Last but not least, a segmentation-based method is proposed to fuse these maps and generate the final decision map. By exploiting the benefits of both the small-scale and large-scale analyses, the segmentation-based multi-scale analysis can lead to a performance leap in forgery localization of CNNs. Numerous experiments are conducted to demonstrate the effectiveness and efficiency of our method.

A Novel Robust Image Forensics Algorithm Based on L1-Norm Estimation

To improve the robustness of the typical image forensics with the noise variance, we propose a novel image forensics approach that based on L1-norm estimation. First, we estimate the kurtosis and the noise variance of the high-pass image. Then, we build a minimum error objective function based on L1-norm estimation to compute the kurtosis and the noise variance of overlapping blocks of the image by an iterative solution. Finally, the spliced regions are exposed through K-means cluster analysis. Since the noise variance of adjacent blocks are similar, our approach can accelerate the iterative process by setting the noise variance of the previous block as the initial value of the current block. According to analytics and experiments, our approach can effectively solve the inaccurate locating problem caused by outliers. It also performs better than reference algorithm in locating spliced regions, especially for those with realistic appearances, and improves the robustness effectively.

Multi-class JPEG Image Steganalysis by Ensemble Linear SVM Classifier

Multi-class steganalysis utilizes multi-class classification methods to predict the category of steganographic schemes used for generating stego files. In this paper we propose a novel multi-class approach towards more efficiently classifying JPEG stego-images with CC-JRM features. Because CC-JRM has successfully cooperates with ensemble classifier in detecting the presence of stego images, we modified ensemble classifier for multi-class steganalysis. The ideas of performing ensemble in different steps results in two schemes in our proposed method. These two schemes are based on different multi-class ensemble strategies, and utilize linear SVM as base classifier. The experimental results shows our methods received better results with less computing cost compared to other multi-class steganalysis method.

Deep-MATEM: TEM query image based cross-modal retrieval for material science literature

With the rapid increasing of published material science literatures, an effective literature retrieving system is important for researchers to obtain relevant information. In this paper we propose a cross-modal material science literatures retrieval method using transmission electron microscopy(TEM) image as query information, which provide a access of using material experiment generated TEM image data to retrieve literatures. In this method, terminologies are extracted and topic distribution are inferred from text part of literatures by using LDA, and we design a multi-task Convolutional Neuron Network(CNN) mapping query TEM image to the relevant terminologies and topic distribution predictions. The ranking score is calculated from output for query image and text data. Experimental results shows our method achieves better performance than multi-label CCA, Deep Semantic Matching(Deep SM) and Modality-Specific Deep Structure(MSDS).