Important Feature Points Research Articles

Application and Optimization of Multi-Factor Authentication and Biometric Technology in Power Grid Energy Network Access Control

With the sudden development in electronic and network technology, the power grid has emerged in several evolved nations and other areas with faster development. Energy storage can enhance the stability, resiliency and efficiency of the electric smart grid. The power grid transfers energy via bidirectional control of energy and information flow. In the power grid, a smart meter is an important component that is deployed on the user side to determine the consumption of energy regularly. However, the data about the usage of electricity leads to the leakage of the private information of the user, which threatens the privacy of the user. The attackers can change the in-transit message induce wrong information and sometimes cause disastrous action. Hence, it is important to assure data security and privacy of users by conducting authentication in power grid communication. An optimization technique named Puzzle Optimization Algorithm (POA) is designed for the efficient privacy-preserving aggregation approach with multi-factor authentication and a biometric scheme. Also, optimal key-based Elliptic Curve Cryptography (ECC) is used to encrypt the sensitive data that are securely stored. Authentication offers high security and texture analysis. In the registration phase, the privacy information of the person with their biometrics is the important feature point of the individual unique identification. Additionally, the designed approach is compared with other existing optimizations to analyse the functionality of the implemented technique. The simulation outcome defined that the suggested technique enhances user privacy and information security than other standard approaches.

Variational autoencoder–based neural electrocardiogram synthesis trained by FEM-based heart simulator

BackgroundFor comprehensive electrocardiogram (ECG) synthesis, a recent promising approach has been based on a heart model with physical and chemical cardiac parameters. However, the problem is that such approach requires a high-cost and limited environment using supercomputers due to the massive computation. ObjectiveThe purpose of this study is to develop an efficient method for synthesizing 12-lead ECG signals from cardiac parameters. MethodsThe proposed method is based on a variational autoencoder (VAE). The encoder and decoder of the VAE are conditioned by the cardiac parameters so that it can model the relationship between the ECG signals and the cardiac parameters. The training data are produced by a comprehensive, Finite Element Method (FEM)-based heart simulator. New ECG signals can then be synthesized by inputting the cardiac parameters into the trained VAE decoder without relying on enormous computational resources. We used two metrics to evaluate the quality of ECG signals synthesized by the proposed model. ResultsExperimental results showed that the proposed model synthesized adequate ECG signals while preserving empirically important feature points and the overall signal shapes. We also explored the optimal model by varying the number of layers and the size of latent variables in the proposed model that balances the model complexity and the simulation accuracy. ConclusionThe proposed method has the potential to become an alternative to computationally-expensive FEM-based heart simulators. It is able to synthesize ECGs from various cardiac parameters within seconds on a personal laptop computer.

Feature descriptors using super-pixels as fuzzy numbers

ABSTRACT The objective of this paper is to extract directly local important region descriptors using image super-pixels and fuzzy numbers. Previous works are based on extracting important feature points like corners in an image then region descriptors are formed around these features. Our novel contribution is to consider directly the most discriminative super-pixels as region descriptors. First, each super-pixel is considered as a fuzzy number. Then the alpha-cut which best represents the fuzzy number is obtained. Finally, according to these alpha-cuts and the cardinality of each fuzzy number the region descriptors are formed. Matching is done according to distances between fuzzy numbers. The Palm-print recognition problem was chosen to show the effectiveness of this approach.

Multimodal predictive classification of Alzheimer's disease based on attention‐combined fusion network: Integrated neuroimaging modalities and medical examination data

AbstractEarly diagnosis of Alzheimer's disease (AD) plays a key role in preventing and responding to this neurodegenerative disease. It has shown that, compared with a single imaging modality‐based classification of AD, synergy exploration among multimodal neuroimages is beneficial for the pathological identification. However, effectively exploiting multimodal information is still a big challenge due to the lack of efficient fusion methods. Herein, a multimodal fusion network based on attention mechanism is proposed, in which magnetic resonance imaging (MRI) and positron emission computed tomography (PET) images are converted into feature vectors with the same dimension, while the demographic information and clinical data are preprocessed and converted into feature vectors through embedding. This attention model can focus on important feature points, fuse the multimodal information more effectively, and thus provide accurate diagnosis and prediction for different pathological stages. The results show that the model achieves an accuracy of 84.1% for triple classification tasks in normal cognition (NC) versus mild cognitive impairment (MCI) versus AD and 93.9% prediction accuracy in stable MCI (sMCI) versus progressive MCI (pMCI). In contrast to the existing multimodal diagnosis methods, our model yields a state‐of‐the‐art accuracy of AD diagnosis, which is powerful and promising in clinical practice.

Antidepressant Drug Design on TCAs and Phenoxyphenylpropylamines Utilizing QSAR and Pharmacophore Modeling.

Depression is a mental illness caused by the imbalance of important neurotransmitters such as serotonin (5-HT) and norepinephrine (NE). It is a serious neurological disorder that could be treated by antidepressant drugs. There are two major classes, such as TCAs and phenoxyphenylpropylamines, which have been proven to be broad-spectrum antidepressant compounds. Several attempts were made to design, synthesize and discover potent antidepressant compounds having the least toxicity and most selectivity towards serotonin and norepinephrine transporters. However, there is hardly any drug design based on quantitative structure-activity relationship (QSAR) and pharmacophore modeling attempted yet. In the present study, many TCAs (dibenzoazepine) and phenoxyphenylpropylamine derivatives are taken into consideration for pharmacophore feature generation followed by pharmacophoric distant related descriptors based QSAR modeling. Furthermore, several five new congeners have been designed which are subjected to the prediction of biological activities in terms of serotonin receptor affinity utilizing validated QSAR models developed by us. An important pharmacophoric feature point C, followed by the generation of a topography of the TCAs and phenoxyphenylpropylamine, has been predicted. The developed pharmacophoric feature-based QSAR can explain 64.2% of the variances of 5-HT receptor antagonism. The best training model has been statistically validated by the prediction of test set compounds. This training model has been used for the prediction of some newly designed congeneric compounds which are comparable with the existed drugs. The newly designed compounds may be proposed for further synthesis and biological screening as antidepressant agents.

An Improved ORB Feature Extraction Algorithm

In this paper, an improved ORB feature extraction algorithm is proposed to solve the insufficient detection effect of the traditional ORB algorithm on the image edges as well as the low actual extraction of feature points. First, the image pyramid layering was established, and then on basis of the improved oFAST12-20 corner point algorithm, the method with a setting dynamic threshold, according to the idea of adaptive threshold, was proposed to extract feature points, and the candidate set of feature points was obtained. Then, Harris algorithm was used to calculate the response value of each candidate point, and the response value was sorted from high to low. Finally, the top 80% of the feature points with high response value were extracted as the most important feature points. The experimental results show that the improved oFAST algorithm improved the extraction rate of feature points and reduces the algorithm time compared with the traditional ORB corner detection.

Hierarchical Learning-Guided human motion quality assessment in big data environment

Image information may be distorted during acquisition, processing, compression, and transmission. It is necessary to propose an intelligent image quality assessment model toward big data environment to quantify the degree of distortion of the image. This paper proposes a quality assessment model for human motion images. In complex scenes, the human body's action posture can be taken as an important feature point. Usually, in different scenes, the parts that affect the quality of the human body's posture are different. In other words, the weights of feature points that affect quality are different in different scenarios. However, due to the categorization of human movements, we can learn the quality assessment methods of different types of movements through sample training. Inspired by feature learning in the field of machine learning, we propose a hierarchical quality learning approach. We cast quality assessment as quality feature learning and layer by layer. The hierarchical quality learning method is based on deep reinforcement learning. The key part is that the method focuses on the region that containing more information on the features of the quality and enlarges the region layer by layer. Finally, we can determine the part of the body that affects the quality assessment. We compare this method with the subjective quality assessment results of the human observers and find that the proposed method achieves effective performance in big data environment to evaluate human motion quality.

Model-Based Coarse-Fine Virtual Calibration and Visual Servo for Augmented Reality-Assisted Peg-in-Hole Microassembly

A three-dimensional (3D)-virtual calibration and visual servo are implemented for augmented reality (AR)-assisted peg-in-hole microassembly operations. By employing 3D model and ray casting, the 3D coordinates on virtual mating rod correspondent to the two-dimensional (2D) virtual image points are extracted. The detecting and tracking of image feature points for calibration is carried out by the proposed algorithm of regional template matching (TM) and scanning with edge fitting (RTM-SEF). For achieving subpixel error between the feature points in real and virtual images, a coarse-fine virtual calibration method is proposed. In regard to the image viewed by the real and virtual cameras, a calibrated virtual camera is utilized to track the mating rod. A visual servo control law including coarse and fine tuning is proposed to ensure sub-pixel error between the most important feature point in the real and virtual images. The AR technology is mainly employed in the alignment between micropeg and mating hole for inserting a micropeg of diameter 80 μm with length 1–1.4 mm into a mating rod with 100 μm hole.

A Novel Face Recognition Algorithm Using Gabor - based KPCA

The Gabor wavelets are used to extract facial features, and then a doubly nonlinear mapping kernel PCA (DKPCA) is proposed to perform feature transformation and face recognition. The conventional kernel PCA nonlinearly maps an input image into a high-dimensional feature space in order to make the mapped features linearly separable. However, this method does not consider the structural characteristics of the face images, and it is difficult to determine which nonlinear mapping is more effective for face recognition. In this work, a new method of nonlinear mapping, which is performed in the original feature space, is defined. The proposed nonlinear mapping not only considers the statistical properties of the input features, but also adopts an Eigen mask to emphasize those important facial feature points The proposed algorithm is evaluated based on the Yale database, the AR database, the ORL database and the YaleB database.

Robust Video Stabilization Using Particle Keypoint Update and l₁-Optimized Camera Path.

Acquisition of stabilized video is an important issue for various type of digital cameras. This paper presents an adaptive camera path estimation method using robust feature detection to remove shaky artifacts in a video. The proposed algorithm consists of three steps: (i) robust feature detection using particle keypoints between adjacent frames; (ii) camera path estimation and smoothing; and (iii) rendering to reconstruct a stabilized video. As a result, the proposed algorithm can estimate the optimal homography by redefining important feature points in the flat region using particle keypoints. In addition, stabilized frames with less holes can be generated from the optimal, adaptive camera path that minimizes a temporal total variation (TV). The proposed video stabilization method is suitable for enhancing the visual quality for various portable cameras and can be applied to robot vision, driving assistant systems, and visual surveillance systems.

Applying riding-posture optimization on bicycle frame design

Customization design is a trend for developing a bicycle in recent years. Thus, the comfort of riding a bike is an important factor that should be paid much attention to while developing a bicycle. From the viewpoint of ergonomics, the concept of “fitting object to the human body” is designed into the bicycle frame in this study. Firstly, the important feature points of riding posture were automatically detected by the image processing method. In the measurement process, the best riding posture was identified experimentally, thus the positions of feature points and joint angles of human body were obtained. Afterwards, according to the measurement data, three key points: the handlebar, the saddle and the crank center, were identified and applied to the frame design of various bicycle types. Lastly, this study further proposed a frame size table for common bicycle types, which is helpful for the designer to design a bicycle.

Key Point Detection in 3D Reconstruction Based On Human-Computer Interaction

Aiming at solving problem of points’ redundancy caused by full automatically detecting points and the problem of large workload caused by picking all points manually, I advanced a new method of picking points which is based on Human-Computer Interaction in our 3D reconstruction platform after automatically detecting points. We first detected and matched points automatically and got the homograph matrix between two images, then projected points which were picked by hand on the one image to the other image, at last we would search the interesting feature points in the neighborhood of corresponding points in the two images. Experiments have shown that this method decreases the redundancy brought by large number of points and successfully finds the important feature points, so it lays a good foundation for 3D reconstruction.

DETECTION OF CROSSOVER & BIFURCATION POINTS ON A RETINAL FUNDUS IMAGE BY ANALYZING NEIGHBORHOOD CONNECTIVITY OF NON-VASCULAR REGIONS AROUND A JUNCTION POINT

Over the last few decades, analysis of retinal vascular images has gained its popularity among the cyber scientists. Retina is unique in nature because of its most important feature, bifurcation & crossover points, serving reliable authentication basis. Using these two points many problem of authentication could be handled easily. Literature has shown that in a retinal vascular structure, bifurcation& crossover points need to be separated for the purpose of authentication. With this motivation in mind, we propose a novel method to segregate vascular bifurcation points from crossover points in any retinal image by analyzing neighborhood connectivity of nonvascular regions around a junction point on retinal blood vessels.

Invariants Feature Points Detection based on Random Sample Estimation

Feature detection is the initial step in any image analysis procedure and is essential for the performance of computer vision applications like stereo vision, object recognition, object tracking systems. Research concerning the detection of feature points for different camera motion images in efficient and fast way. In this work, techniques of corner detection, geometric moments and random sampling are presented to simply and accurately locate the important feature points in image. For each extracted feature in image, a descriptor is calculated and based on the homograph transformation the matching is done. The results of experiments conducted on images taken by handheld camera and compared with the most famous SIFT method. The results show that the proposed algorithm is accurate, fast, efficient and robust under noise, transformation and compression circumstances.

Enhanced Gabor Feature Based Classification Using a Regularized Locally Tensor Discriminant Model for Multiview Gait Recognition

This paper presents a novel multiview gait recognition method that combines the enhanced Gabor (EG) representation of the gait energy image and the regularized local tensor discriminant analysis (RLTDA) method. EG first derives desirable gait features characterized by spatial frequency, spatial locality, and orientation selectivity to cope with the variations due to surface, shoe types, clothing, carrying conditions, and so on. Unlike traditional Gabor transformation, which does not consider the structural characteristics of the gait features, our representation method not only considers the statistical property of the input features but also adopts a nonlinear mapping to emphasize those important feature points. The dimensionality of the derivation of EG gait feature is further reduced by using RLTDA, which directly obtains a set of locally optimal tensor eigenvectors and can capture nonlinear manifolds of gait features that exhibit appearance changes due to variable viewing angles. An aggregation scheme is adopted to combine the complementary information from differently RLTDA recognizers at the matching score level. The proposed method achieves the best average Rank-1 recognition rates for multiview gait recognition based on image sequences from the USF HumanID gait challenge database and the CASIA gait database.

최인접 거리 비율 정합을 이용한 영상 특징점 선택 방법

본 논문에서는 현재 진행 중인 MPEG(Motion Picture Experts Group, ISO/IEC JTC1 SC29 WG11)의 표준화 작업 중 CDVS(Compact Descriptor for Visual Search)의 CE-7(Core Experiment)인 특징점 선택에 대한 방법을 제안한다. 서술자의 경량화를 위해서는 영상으로부터 추출된 많은 수의 특징점들 중에서 영상 정합에 사용될 중요한 특징점들을 선택해야 한다. 본 논문에서는 최 인접 거리 비율 정합(Nearest Neighbor distance ratio matching) 방법에 의해 영상 정합 단계에서 사용되지 않고 버려지는 특징점들을 미리 추출 단에서 제거하는 방법 제안하였다. 제안된 방법을 통하여 적은 비트 전송률을 요하는 시스템에서 특징점의 낭비를 피할 수 있고 결과적으로 추가적인 특징점을 사용할 수 있으므로 전체적인 성능 향상을 얻을 수 있었다. 제안된 알고리즘을 통하여 Pair-wise 정합 실험에서 기존의 Test Model 대비 최고 2.3%의 성공율(True positive rate)의 향상을 보였다. In this paper, we propose a feature point selection method for MPEG CDVS CE-7 which is processing on International Standard task. Among a large number of extracted feature points, more important feature points which is used in image matching should be selected for the compactness of image descriptor. The proposed method is that remove the feature point in the extraction phase which is filtered by nearest neighbor distance ratio matching in the matching phase. We can avoid the waste of the feature point and employ additional feature points by the proposed method. The experimental results show that our proposed method can obtain true positive rate improvement about 2.3% in pair-wise matching test compared with Test Model.

Survey on Various Gesture Recognition Techniques for Interfacing Machines Based on Ambient Intelligence

Gesture recognition is mainly apprehensive on analyzing the functionality of human wits. The main goal of gesture recognition is to create a system which can recognize specific human gestures and use them to convey information or for device control. Hand gestures provide a separate complementary modality to speech for expressing ones ideas. Information associated with hand gestures in a conversation is degree,discourse structure, spatial and temporal structure. The approaches present can be mainly divided into Data-Glove Based and Vision Based approaches. An important face feature point is the nose tip. Since nose is the highest protruding point from the face. Besides that, it is not affected by facial expressions.Another important function of the nose is that it is able to indicate the head pose. Knowledge of the nose location will enable us to align an unknown 3D face with those in a face database. Eye detection is divided into eye position detection and eye contour detection. Existing works in eye detection can be classified into two major categories: traditional image-based passive approaches and the active IR based approaches. The former uses intensity and shape of eyes for detection and the latter works on the assumption that eyes have a reflection under near IR illumination and produce bright/dark pupil effect. The traditional methods can be broadly classified into three categories: template based methods,appearance based methods and feature based methods. The purpose of this paper is to compare various human Gesture recognition systems for interfacing machines directly to human wits without any corporeal media in an ambient environment.

Determination of 3D necklines from scanned human bodies

Accurate recognition of the human body is an essential procedure of clothing pattern making, garment fit evaluation, and sizing survey. Over the years, many research efforts have been devoted to define and/or identify features (points and lines) of a human body. However, the neckline, one imperative feature of a human body is still difficult to accurately identify. It is complicated to determine a neckline geometrically due to the large variety of body figures, and the defined neckline must fulfill the requirements of apparel manufacturing and textile properties. The aim of the paper is to propose a new and practical method to identify the neckline from 3D digitized bodies. Firstly, a torso represented by piecewise B-spline curves is generated from a triangulated human mesh model. Secondly, important feature points including front neck point (FNP), back neck point (BNP) and side neck point (SNP) are identified. A cutting-surface is defined and the profile of such cutting surface is construed by those identified feature points. By intersecting the cutting-surface with the piecewise B-spline curve torso, a neckline is then properly determined. The proposed method is verified to be effective for generating necklines for human subjects with varied neck silhouettes. A neckline fitting survey through real subjects’ wear trial is conducted to evaluate and compare the proposed method with the traditional methods.

Augmented DT-CWT feature based classification using Regularized Neighborhood Projection Discriminant Analysis for face recognition

This paper presents a novel face recognition method which integrates the Augmented Dual-Tree Complex Wavelet Transform (ADT-CWT) representation of face images and Regularized Neighborhood Projection Discriminant Analysis (RNPDA) method. ADT-CWT first derives desirable facial features characterized by spatial frequency, spatial locality, and orientation selectivity to cope with the variations due to illumination and facial expression changes. Different from DT-CWT, which does not consider the structural characteristics of the face images, our representation method not only considers the statistical property of the input features but also adopts an Eigenmask to emphasize those important facial feature points. The dimensionality of the derivation of ADT-CWT feature is further reduced by using RNPDA, which directly obtain a set of optimal eigenvectors with a simple regression framework and thus can overcome the small sample size problem of NPDA. Extensive experiments have been made to compare the recognition performance of the proposed method with some popular dimensionality reduction methods on the FERET database, the extended YALEB database and the CMU PIE database. The results verify the effectiveness of the proposed method.

FACE RECOGNITION USING FUZZY INFERENCE SYSTEM

The face is the feature which distinguishes a person. Facial appearance is vital for human recognition. It has certain features like forehead, skin, eyes, ears, nose, cheeks, mouth, lip, teeth etc which helps us, humans, to recognize a particular face from millions of faces even after a large span of time and despite large changes in their appearance due to ageing, expression, viewing conditions and distractions such as disfigurement of face, scars, beard or hair style. A face is not merely a set of facial features but is rather but is rather something meaningful in its form.In this paper, depending on the various facial features, a system is designed to recognize them. To reveal the outline of the face, eyes, ears, nose, teeth etc different edge detection techniques have been used. These features are extracted in the term of distance between important feature points. The feature set obtained is then normalized and are feed to artificial neural networks so as to train them for reorganization of facial images.