Proposed Thresholding Scheme Research Articles

Recognition of Cosmic Ray Images Obtained from CMOS Sensors Used in Mobile Phones by Approximation of Uncertain Class Assignment with Deep Convolutional Neural Network.

In this paper, we describe the convolutional neural network (CNN)-based approach to the problems of categorization and artefact reduction of cosmic ray images obtained from CMOS sensors used in mobile phones. As artefacts, we understand all images that cannot be attributed to particles’ passage through sensor but rather result from the deficiencies of the registration procedure. The proposed deep neural network is composed of a pretrained CNN and neural-network-based approximator, which models the uncertainty of image class assignment. The network was trained using a transfer learning approach with a mean squared error loss function. We evaluated our approach on a data set containing 2350 images labelled by five judges. The most accurate results were obtained using the VGG16 CNN architecture; the recognition rate (RR) was 85.79% ± 2.24% with a mean squared error (MSE) of 0.03 ± 0.00. After applying the proposed threshold scheme to eliminate less probable class assignments, we obtained a RR of 96.95% ± 1.38% for a threshold of 0.9, which left about 62.60% ± 2.88% of the overall data. Importantly, the research and results presented in this paper are part of the pioneering field of the application of citizen science in the recognition of cosmic rays and, to the best of our knowledge, this analysis is performed on the largest freely available cosmic ray hit dataset.

Threshold-Based Post-Quantum Secure Verifiable Multi-Secret Sharing for Distributed Storage Blockchain

Blockchain systems store transaction data in the form of a distributed ledger where each node stores a copy of all data, which gives rise to storage issues. It is well-known that the tremendous storage and distribution of the block data are common problems in blockchain systems. In the literature, some types of secret sharing schemes are employed to overcome these problems. The secret sharing method is one of the most significant cryptographic protocols used to ensure the privacy of the data. The main purpose of this paper is to improve the recent distributed storage blockchain systems by proposing an alternative secret sharing method. We first propose a secure threshold verifiable multi-secret sharing scheme that has the verification and private communication steps based on post-quantum lattice-based hard problems. We then apply the proposed threshold scheme to the distributed storage blockchain (DSB) system to share transaction data at each block. In the proposed DSB system, we encrypt the data block with the AES-256 encryption algorithm before distributing it among nodes at each block, and both its secret key and the hash value of the block are privately shared among nodes simultaneously by the proposed scheme. Thereafter, in the DSB system, the encrypted data block is encoded by the Reed–Solomon code, and it is shared among nodes. We finally analyze the storage and recovery communication costs and the robustness of the proposed DSB system. We observe that our approach improves effectively the recovery communication cost and makes it more robust compared to the previous DSB systems. It also improves extremely the storage cost of the traditional blockchain systems. Furthermore, the proposed scheme brings to the DSB system the desirable properties such as verification process and secret communication without private channels in addition to the known properties of the schemes used in the previous DSB systems. As a result of the flexibility on the threshold parameter of the scheme, a diverse range of qualified subsets of nodes in the DSB system can privately recover the secret values.

Image enhancement with the preservation of brightness and structures by employing contrast limited dynamic quadri-histogram equalization

Image enhancement has been widely applied to medical science, industry, agriculture, and military, for which it is very important to preserve brightness and structures. In order to overcome the shortcomings of over-enhancement and over-smoothing of traditional histogram equalization methods, this paper proposes an effective image enhancement strategy named as contrast limited dynamic quadri-histogram equalization (CLDQHE) which includes three steps: (1) the original histogram of an image is divided by a proposed threshold scheme into four sub-histograms. (2) An adaptive histogram clipping method is proposed to control the enhancement rate, which can adjust the clipping threshold. (3) Each sub-histogram is mapped to a new dynamic range and is equalized independently. Both quantitatively and qualitatively experimental results demonstrate that the proposed CLDQHE method can yield pleasing results with the preservation of brightness and structures, and even outperforms the state-of-the-arts.

Emotion Recognition with Refined Labels for Deep Learning.

The traditional emotion classification framework usually fits all the features segments of the same trial to a fixed annotation. Considering the fact that emotion is a reaction to stimuli that lasts for varied periods, we argue that the indiscriminate annotation is equivalent to taking the emotional state as fixed within the whole trial, leading to a decrease of the classification accuracy. In this study, we attempt to alleviate this issue by developing a thresholding scheme, converting the continuous emotional trace into a three-class annotation temporally. The features within a trial are therefore assigned to varied emotional states, resulting in an improvement in the accuracy. A long short term memory (LSTM) networks-based emotion classification framework is implemented, to which the proposed thresholding scheme is applied. A subset of MAHNOB-HCI dataset with continuous emotional annotation is used. The EEG signal and frontal facial video are used for feature extraction. The experiment results demonstrate that the proposed scheme provides statistically significant improvement to the three-class classification accuracy of the EEG feature-based LSTM network (p-value = 0.0329).

Secure rational numbers equivalence test based on threshold cryptosystem with rational numbers

In this study, we consider an equivalence test on rational numbers in a scenario with K+2 distributed parties, Alice, Bob, P1,P2,…,PK, where Alice has a private rational number xa, Bob has a private rational number xb, and party Pi has a secret si, where i∈{1,2,…,K},K ≥ 2. The parties want to cooperatively detect whether xa=xb without revealing any information about their secrets. This problem has many applications in online collaboration, such as e-voting, which requires public verifiability. First, we develop a provably secure threshold cryptosystem for rational numbers. Next, based on the proposed threshold scheme, we construct a distributed plaintext equivalence test protocol in an honest majority environment. We prove that the proposed protocol is secure and robust in the standard (ideal/real) model.

Thresholding Scheme Based on Boundary Pixels of Stripes for Visible Light Communication With Mobile-Phone Camera

In mobile-phone camera-based visible light communication (VLC) system, column matrix selection and threshold setting are very important steps for signal demodulation. As the blooming effect of complementary metal-oxide-semiconductor camera fluctuates grayscale values of pixels, thus causing a decrease in the accuracy of distinguishing data logic 0 and 1. Facing grayscale value fluctuation, using traditional polynomial fitting as the threshold to distinguish data logic 0 and 1, the achievable system performance is not very good. This limits applications of the VLC system. Therefore, advanced thresholding schemes are needed to increase the accuracy of distinguishing data logic 0 and 1, thereby improving the system performance. In this paper, we propose and experimentally demonstrate for the first time a thresholding scheme based on boundary pixels of stripes for improving the VLC system performance. First, we describe the process of boundary pixel position acquisition for the thresholding scheme in detail. Then, we analyze and discuss the selection of the optimum column matrix for signal demodulation. Finally, we demonstrate experiments to verify the feasibility of our proposed thresholding scheme. Experimental results show that our proposed thresholding scheme can achieve a good performance even when grayscale values fluctuate greatly. Compared with the conventional third-order polynomial fitting scheme, our proposed thresholding scheme has advantages in both of transmission distance and data rate. With aid of the optimum selected column matrix and our proposed thresholding scheme, a data rate of 4.5 Kb/s (2.5 Kb/s) is successfully achieved with the transmission distance of 50 cm (175 cm) under the 7% forward error correction limit (bit-error-rate = $3.8\times 10^{-3}$ ) for the VLC system with mobile-phone camera.

A thresholding scheme of eliminating false detections on vehicles in wide-area aerial imagery

Post-processings are usually necessary to reduce false detections on vehicles in wide-area aerial imagery. In order to improve the performance of vehicle detection, we propose a two-stage scheme, which consists of a thresholding method by constructing a pixel-weight based thresholding policy to classify pixels in the greyscale feature map of an automatic detection algorithm followed by morphological filtering. We use two aerial videos for performance evaluation, and compare the automatic detection results with the ground-truth objects. We compute average F-score and percentage of wrong classifications towards six detection algorithms before and after applying the proposed scheme. We measure the variation of overlap ratios from detections to objects, and establish sensitivity analysis to evaluate the performance of proposed scheme by combining it on each of two representative algorithms. Simulation results verify both validity and efficiency of the proposed thresholding scheme, also display the difference of detection performance between datasets and among algorithms.

Topological Filtering of Dynamic Functional Brain Networks Unfolds Informative Chronnectomics: A Novel Data-Driven Thresholding Scheme Based on Orthogonal Minimal Spanning Trees (OMSTs).

The human brain is a large-scale system of functionally connected brain regions. This system can be modeled as a network, or graph, by dividing the brain into a set of regions, or “nodes,” and quantifying the strength of the connections between nodes, or “edges,” as the temporal correlation in their patterns of activity. Network analysis, a part of graph theory, provides a set of summary statistics that can be used to describe complex brain networks in a meaningful way. The large-scale organization of the brain has features of complex networks that can be quantified using network measures from graph theory. The adaptation of both bivariate (mutual information) and multivariate (Granger causality) connectivity estimators to quantify the synchronization between multichannel recordings yields a fully connected, weighted, (a)symmetric functional connectivity graph (FCG), representing the associations among all brain areas. The aforementioned procedure leads to an extremely dense network of tens up to a few hundreds of weights. Therefore, this FCG must be filtered out so that the “true” connectivity pattern can emerge. Here, we compared a large number of well-known topological thresholding techniques with the novel proposed data-driven scheme based on orthogonal minimal spanning trees (OMSTs). OMSTs filter brain connectivity networks based on the optimization between the global efficiency of the network and the cost preserving its wiring. We demonstrated the proposed method in a large EEG database (N = 101 subjects) with eyes-open (EO) and eyes-closed (EC) tasks by adopting a time-varying approach with the main goal to extract features that can totally distinguish each subject from the rest of the set. Additionally, the reliability of the proposed scheme was estimated in a second case study of fMRI resting-state activity with multiple scans. Our results demonstrated clearly that the proposed thresholding scheme outperformed a large list of thresholding schemes based on the recognition accuracy of each subject compared to the rest of the cohort (EEG). Additionally, the reliability of the network metrics based on the fMRI static networks was improved based on the proposed topological filtering scheme. Overall, the proposed algorithm could be used across neuroimaging and multimodal studies as a common computationally efficient standardized tool for a great number of neuroscientists and physicists working on numerous of projects.

A Distributed and Collaborative Intrusion Detection Architecture for Wireless Mesh Networks

Wireless Mesh Network (WMN) is an emerging heterogeneous network architecture that is growing in importance among traditional wireless communication systems as a cost-effective way of providing Internet services. However, WMNs are particularly vulnerable to malicious nodes given their inherent attributes such as decentralized infrastructure and high dependence of node cooperation. We then propose a distributed and Collaborative Intrusion Detection System (CIDS) architecture for detecting insider attacks at real-time, which comprises: i) a Routing Protocol Analyzer (RPA) to analyze the collected routing traffic and generate respective Routing Events; ii) a Distributed Intrusion Detection Engine (DIDE) that treats the Routing Events by applying Routing Constraints and calculate related Misbehaving Metrics; iii) a Cooperative Consensus Mechanism (CCM) to check the Misbehaving Metrics using a proposed threshold scheme and to track down the source of intrusion. The entire CIDS solution is implemented in a virtualized mesh network platform. The experimental results show the proposed CIDS architecture efficiently detects message fabrication attacks with good precision and low resource consumption.

Optimum Threshold for Indoor UWB ToA-Based Ranging

This paper proposes a method for setting the threshold for ultra-wide-band (UWB) threshold-based ranging in indoor scenarios. The optimum threshold is derived based on the full analysis of the ranging error, which is equivalent to the probability of correct detection of first arriving signal in time-based ranging techniques. It is shown that the probability of correct detection is a function of first arriving signal, which has variations with two independent distributions. On the one hand, the first arriving signal varies in different positions with the same range due to multipath interference and on the other, it is a function of distance due to free space path-loss. These two distributions are considered in the derivation of the ranging error, based on which the optimum threshold is obtained. A practical method to derive this threshold is introduced based on the standard channel model. Extensive Monte Carlo simulations, ray-tracing simulations and ranging measurements confirm the analysis and the superior performance of the proposed threshold scheme.

Multiplicative Forward-Secure Threshold Signature Scheme

The devastating consequence of secret key exposure in digital signature is that any signature can be forged and cannot be trusted. To mitigate the damage of secret key exposure, forward-secure signature schemes and threshold signature schemes are devised. In this paper, we propose a robust forward-secure threshold signature scheme with the applicability to mobile ad-hoc network in mind. Our main objective is to reduce interaction among the set of signers and to reduce the dependency on broadcast as well as private point-to-point connections. We achieve this by avoiding the regular polynomial sharing and employing multiplicative sharing in a threshold structure. The security of our proposed threshold scheme is reducible to the security of a single-user scheme, which has been proven secure under the random oracle model.

Error-reduction techniques and error analysis for fully phase- and amplitude-based encryption

The performance of fully phase- and amplitude-based encryption processors is analyzed. The effects of noise perturbations on the encrypted information are considered. A thresholding method of decryption that further reduces the mean-squared error (MSE) for the fully phase- and amplitude-based encryption processes is provided. The proposed thresholding scheme significantly improves the performance of fully phase- and amplitude-based encryption, as measured by the MSE metric. We obtain analytical MSE bounds when thresholding is used for both decryption methods, and we also present computer-simulation results. These results show that the fully phase-based method is more robust. We also give a formal proof of a conjecture about the decrypted distribution of distorted encrypted information. This allows the analytical bounds of the MSE to be extended to more general non-Gaussian, nonadditive, nonstationary distortions. Computer simulations support this extension.