Image-based Scheme Research Articles

A novel image-based scheme for automated detection and identification of weld defects

A novel image-based scheme for automated detection and identification of weld defects

Wrist Torque Estimation via Electromyographic Motor Unit Decomposition and Image Reconstruction.

Neural interface using decomposed motor units (MUs) from surface electromyography (sEMG) has allowed non-invasive access to the neural control signals, and provided a novel approach for intuitive human-machine interaction. However, most of the existing methods based on decomposed MUs merely adopted the discharge rate (DR) as the feature representations, which may lack local information around the discharge instant and ignore the subtle interactions of different MUs. In this study, we proposed an MU-specific image-based scheme for wrist torque estimation. Specifically, the high-density sEMG signals were decoded into motor unit spike trains (MUSTs), and then MU-specific images were reconstructed with MUSTs and corresponding motor unit action potential (MUAP). A convolutional neural network was used to learn representative features from MU-specific images automatically, and further to estimate wrist torques. The results demonstrated that the proposed method outperformed three conventional and a deep-learning regression approaches using DR features, with the estimation accuracy R2 of 0.82 ± 0.09, 0.89 ± 0.06, and nRMSE of 12.6 ± 2.5%, 11.0 ± 3.1% for pronation/supination and flexion/extension, respectively. Further, the analysis of the extracted features from MU-specific images showed a higher correlation than DR for recorded torques, indicating the effectiveness of the proposed method. The outcomes of this study provide a novel and promising perspective for the intuitive control of neural interfacing.

A Security Analysis of Captchas with Large Character Sets

Captcha, which can prevent computer programs from attacking websites, has been the most important security technology for many years. The most popularly deployed Captcha is the text-based scheme. The vast majority of the existing text Captchas are designed with English letters and Arabic numerals. Recently, text Captchas with large character sets are being increasingly popular. From the perspective of attackers, larger character set means greater solution space and better theoretical security. However, the security of Captchas with large character sets in real world has never been studied comprehensively. In this article, we introduce a simple, fast, and effective deep learning method to attack these newly emerging Captchas. Taking 11 Chinese Captchas as representatives, we ran our experimental attack on each of them. Our attack achieved high success rates, ranging from 34.7 to 86.9 percent at an average speed of 0.175 seconds on these schemes. All of the results show that the Chinese text Captcha can be easily broken, demonstrating that text Captchas with large character sets are also insecure in existing forms. As a substitute, we proposed a 3D image-based scheme combining semantic comprehension and dragging action. The preliminary experimental results show that it is more robust than current text-based schemes.

The Image-Based Positioning Technique Using Inter-Pixel Relation Network

Accurate location information is very important for human beings and agents. However, GPS has very limited positioning accuracy or even can not work in challenging environments such as indoor and urban canyons, and other solutions suffer from different shortcomings. It is very meaningful to use the scene image captured by the high-definition camera on the smart phone to locate indoor through the neural network, but the existing image-based positioning methods do not take into account the interrelationship of diffenrent instances of the same feature space. In this paper we present a novel network architecture, which embeds Scene Enhancement module and Relational Network (RN) into the end-to-end positioning system. The system was tested in three typical scenarios, and the test results show that its performance outerperforms existing image-based schemes in most cases.

An efficient image-based verification scheme by fusion of double random phase encoding and dynamic chaotic map

This paper introduces an efficient image-based verification approach that integrates double random phase encoding (DRPE) and chaotic dynamic mapping. The main objective is to propose a new DRPE-based image verification scheme by replacing random phase masks (RPM) with efficient chaotic phase masks (CPM) which are dependent on an input image. For verification, the input image is first encrypted using the DRPE algorithm with the CPMs generated by our proposed methods. Then, the encrypted input image is sent to a database for verification. The peak-to-correlation energy between the encrypted input image and the encrypted reference image in the database verifies the image authenticity. The experiments reveal that the proposed scheme is highly sensitive to the key changes and represents good resistance against statistical attacks. This method can also be applied in the area of identity verification.

A spatiotemporal outlier detection method based on partial least squares discriminant analysis and area Delaunay triangulation for image-based process monitoring

ABSTRACTOver the past two decades, statistical process control has evolved from monitoring individual data points to linear profiles to image data. Image sensors are now being deployed in complex systems at increasing rates due to the rich information they can provide. As a result, image data play an important role in process monitoring in different application domains ranging from manufacturing to service systems. Many of the existing process monitoring methods fail to take full advantage of the image data due to the data's complex nature in both the spatial and temporal domains. This article proposes a spatiotemporal outlier detection method based on the partial least squares discriminant analysis and a control statistic based on the area Delaunay triangulation of the squared prediction errors to improve the performance of an image-based monitoring scheme. First, the discriminant analysis of the partial least squares is used to efficiently extract the most important features from the high-dimensional image data to identify the benchmark images of the products and obtain the pixel value errors. Next, the squared errors resulting from the previous step are connected using a Delaunay triangulation to form a surface, the area of which is used as the control statistic for the purpose of outlier detection. A real case study at a paper product manufacturing company is used to compare the performance of the proposed method in detecting different types of outliers with some of the existing methods and demonstrate the merit of the proposed method.

Novel Image Set Compression Algorithm Using Rate-Distortion Optimized Multiple Reference Image Selection

Image set compression has recently become an active research topic due to the explosion of digital photographs. In order to efficiently compress the image sets of similar images with moving objects, in this paper, we propose a novel algorithm for image set compression using multiple reference images. First, for an image set, its depth-constrained minimum arborescence is generated. We then present a reference image candidate determination method to build the reference image candidates for the images of the set. Furthermore, we propose a rate-distortion optimized multiple reference image selection method. This method compares the correlation between every image and each of its reference image candidates to produce its multiple reference images. Finally, compressed image data are achieved by employing block-based motion compensation and residue coding. In addition, we also give a new way of access to images to keep the same access delay with single reference image-based schemes. Compared with the state-of-the-art image compression algorithms, experimental results show that our proposed algorithm can significantly improve the image compression performance.

Toward Improved Audio CAPTCHAs Based on Auditory Perception and Language Understanding

A so-called completely automated public Turing test to tell computers and humans apart (CAPTCHA) represents a challenge-response test that is widely used on the Internet to distinguish human users from fraudulent computer programs, often referred to as bots. To enable access for visually impaired users, most Web sites utilize audio CAPTCHAs in addition to a conventional image-based scheme. Recent research has shown that most currently available audio CAPTCHAs are insecure, as they can be broken by means of machine learning at relatively low costs. Moreover, most audio CAPTCHAs suffer from low human success rates that arise from severe signal distortions. This article proposes two different audio CAPTCHA schemes that systematically exploit differences between humans and computers in terms of auditory perception and language understanding, yielding a better trade-off between usability and security as compared to currently available schemes. Furthermore, we provide an elaborate analysis of Google’s prominent reCAPTCHA that serves as a baseline setting when evaluating our proposed CAPTCHA designs.

A Comparative Study of Authentication Schemes with Security and Usability of IPAS

User authentication is one of the most important topics in information security. Conventional user strong-password authentication schemes have the common drawback that user has to memorize a hard-to-remember textual password, and therefore their applications are restricted. As an alternative, token and biometric-based authentication systems were introduced. However, they have not improved substantially to justify the investment. Thus, an image-based authentication scheme has been proposed as a possible alternative solution to text-based authentication, motivated particularly by the fact that humans can remember images better than text. Click based graphical passwords that use background images do not have a sufficiently large password space suited for most Internet applications. Also, it suffered due to shoulder-surfing and replay attacks. All graphical and image algorithms comprise two different aspects: security and usability. Unfortunately, none of the graphical algorithms are able to cover both aspects at the same time. In this paper, we evaluate the usability and security of different authentication schemes and compare them with our proposed scheme, which is the Implicit Password Authentication System (IPAS) by an experiment and a questionnaire survey.

Uncalibrated Neuro-Visual Servoing Control for Multiple Robot Arms

Diverse image-based tracking schemes for a robot moving in free motion have been proposed and experimentally validated. However, few visual servoing schemes have addressed the tracking of the desired trajectory and the contact forces for multiple robot arms. The main difficulty stems from the fact that camera information cannot be used to drive force trajectories. Recognizing this fact, a unique error manifold that includes position-velocity and force errors in orthogonal complements is proposed. A synergistic scheme that fuses camera, encoder and force sensor signals into a unique error manifold allows proposing a control system which guarantees exponential tracking errors under parametric uncertainty. Additionally a small neural network driven by a second order sliding mode surface is derived to compensate robot dynamics. Residual errors that arise because of the finite size of the neural network are compensated via an orthogonalized second order sliding mode. The performance of the proposed scheme, in two significant applications of the multiple robot arms, is validated through numerical simulations.

Respiration-induced movement correlation for synchronous noninvasive renal cancer surgery

Noninvasive surgery (NIS), such as high-intensity focused ultrasound (HIFU)-based ablation or radiosurgery, is used for treating tumors and cancers in various parts of the body. The soft tissue targets (usually organs) deform and move as a result of physiological processes such as respiration. Moreover, other deformations induced during surgery by changes in patient position, changes in physical properties caused by repeated exposures and uncertainties resulting from cavitation also occur. In this paper, we present a correlation-based movement prediction technique to address respiration-induced movement of the urological organs while targeting through extracorporeal trans-abdominal route access. Among other organs, kidneys are worst affected during respiratory cycles, with significant three-dimensional displacements observed on the order of 20 mm. Remote access to renal targets such as renal carcinomas and cysts during noninvasive surgery, therefore, requires a tightly controlled real-time motion tracking and quantitative estimate for compensation routine to synchronize the energy source(s) for precise energy delivery to the intended regions. The correlation model finds a mapping between the movement patterns of external skin markers placed on the abdominal access window and the internal movement of the targeted kidney. The coarse estimate of position is then fine-tuned using the Adaptive Neuro-Fuzzy Inference System (ANFIS), thereby achieving a nonlinear mapping. The technical issues involved in this tracking scheme are threefold: the model must have sufficient accuracy in mapping the movement pattern; there must be an image-based tracking scheme to provide the organ position within allowable system latency; and the processing delay resulting from modeling and tracking must be within the achievable prediction horizon to accommodate the latency in the therapeutic delivery system. The concept was tested on ultrasound image sequences collected from 20 healthy volunteers. The results indicate that the modeling technique can be practically integrated into an image-guided noninvasive robotic surgical system with an indicative targeting accuracy of more than 94%. A comparative analysis showed the superiority of this technique over conventional linear mapping and modelfree blind search techniques.

An Image-Based CAPTCHA Scheme Exploiting Human Appearance Characteristics

CAPTCHAs are automated tests that are there to avoid misuse of computing and information resources by bots. Typical text-based CAPTCHAs are proven to be vulnerable against malicious automated programs. In this paper, we present an image-based CAPTCHA scheme using easily identifiable human appearance characteristics that overcomes the weaknesses of current text-based schemes. We propose and evaluate two applications for our scheme involving 25 participants. Both applications use same characteristics but different classes against those characteristics. Application 1 is optimized for security while application 2 is optimized for usability. Experimental evaluation shows promising results having 83% human success rate with Application 2 as compared to 62% with Application 1.

Visual servoing for path reaching with nonholonomic robots

SUMMARYWe present two visual servoing controllers (pose-based and image-based) enabling mobile robots with a fixed pinhole camera to reach and follow a continuous path drawn on the ground. The first contribution is the theoretical and experimental comparison between pose-based and image-based techniques for a nonholonomic robot task. Moreover, our controllers are appropriate not only for path following, but also for path reaching, a problem that has been rarely tackled in the past. Finally, in contrast with most works, which require the path geometric model, only two path features are necessary in our image-based scheme and three in the pose-based scheme. For both controllers, a convergence analysis is carried out, and the performance is validated by simulations, and outdoor experiments on a car-like robot.

Chaotic based reconstructed phase space features for detecting ventricular fibrillation

Among the variety of cardiac arrhythmias, ventricular fibrillation (VF) and ventricular tachycardia (VT) are life-threatening; thus, accurate classification of these arrhythmias is a crucial task for cardiologists. Nevertheless, VT and VF signals are very similar in the time domain and accurate distinguishing these signals with naked eyes in some cases is impossible. In this paper, a novel self-similarity image-based scheme is introduced to classify the underlying information of VT, VF and normal electrocardiogram (ECG) signals. In this study, VT, VF and normal ECG signals are selected from CCU of the Royal Infirmary of Edinburgh and MIT-BIH datasets. According to the time delay method, signal samples can be assigned to state variables and a trajectory can be achieved. To extract the proposed self-similarity feature, first, two different trajectories from each signal trial are drawn according to two different delay time values. The two-dimensional state space of each trial trajectory is considered as an image. Therefore, two trajectory images are produced for each signal. Number of visited pixels in the first image is determined and is subtracted from that of the second image as the self-similarity feature of that signal. Moreover, another scheme is proposed to have a better estimation of self-similarity in which the logical AND operator is applied to both images (matrices) of each ECG trial. The third proposed criterion is similar to box counting method by this difference that each pixel is assigned a weight according to the trajectory density at that point and finally visited weighted pixels are counted. To classify VF from VT and normal ECG, a threshold is determined through the cross validation phase under the Receiver Operating Characteristic (ROC) criterion. To assess the proposed methods, the mentioned signals are classified using the-state-of-art chaotic features such as correlation dimension, the largest Lyapunov exponent and Approximate Entropy (ApEn). Experimental results indicate superiority of the proposed method in classifying the VT, VF and normal ECG signals compared to present traditional schemes. In addition, computational complexity of the introduced methods is very low and can be implemented in real-time applications.

C2 月小型実験機SLIMの画像航法(月探査(1))

The ability of pinpoint soft landing to lunar or planetary surfaces can broaden the range of scientific missions. SLIM (Smart Lander for Investigating Moon) is a small spacecraft (〜400 kg) for demonstration of the pinpoint landing technology and is being studied extensively in ISAS, JAXA. The key technology for SLIM is image-based navigation. The landing strategy consists of 'powered descent' with continuous burning of the main thrusters and 'vertical descent' below an altitude of about 3.5 km. In the powered descent, the spacecraft is controlled by surface relative navigation with image-based schemes such as geographical feature (e.g., crater) matching. In the vertical descent, autonomous obstacle detection and avoidance is carried out using image-based information like shadow. In this paper, image processing methods under consideration and the concept of hardware implementation are described.