Direction Histogram Research Articles

Writer identification using redundant writing patterns and dual-factor analysis of variance

Writer identification (WI) is a typical pattern recognition problem with the goal of recognizing the writer of a text from images of his or her handwriting. For handwriting-based applications, a new approach is required that can confirm the writer based on a very small amount of available text. However, due to the limited number of characters and the diverse contents of the available samples, it is still challenging to achieve a high recognition rate on large-scale databases. To solve this problem, this paper proposes an efficient method for offline text-independent WI using redundant writing patterns and dual-factor analysis of variance (DF-ANOVA). In this method, a search for writing patterns in a limited number of handwritten texts is first conducted to generate meaningful codebooks. Then, these writing patterns are described by means of the simplified Wigner distribution function (WDF) and improved directional index histogram (DIH) descriptors to extract both structural and texture features, and the two corresponding feature distances are integrated by means of a fuzzy integral rule to make the final decision. The Fisher discriminant ratio (FDR) is used for feature selection, and DF-ANOVA is used to eliminate the influence of factors associated with the labels of samples on the feature distance. The results of evaluations on the IAM (96.92%), Firemaker (96.4%), and Uyghur2016 (100%) databases illustrate that this system shows strong robustness to an increasing number of writers and a decreasing number of words in the sample texts.

HDG and HDGG: an extensible feature extraction descriptor for effective face and facial expressions recognition

The potential of facial and facial expression recognitions has gained increased interest in social interactions and biometric identification. Earlier facial identification methods suffer from drawbacks due to the lower identification accuracy under difficult lighting conditions. This paper presents two novel new descriptors called Histogram of Directional Gradient (HDG) and Histogram of Directional Gradient Generalized (HDGG) to extracting discriminant facial expression features for better classification accuracy with good efficiency than existing classifiers. The proposed descriptors are based on the directional local gradients combined with SVM (Support Vector Machine) linear classification. To build an efficient face and facial expression recognition, features with reduced dimension are used to boost the performance of the classification. Experiments are conducted on two public-domain datasets: JAFFE for facial expression recognition and YALE for face recognition. The experiment results show the best overall accuracy of 92.12% compared to other existing works. It demonstrates a fast execution time for face recognition ranging from 0.4 to 0.7 s in all evaluated databases.

Research on Statistics of Passenger Transfer Flow at Urban Rail Transit Station

With the continuous advancement of urbanization, the urban rail transit network continues to be dense, and the number of transfer stations increases, resulting in an increase in the flow of interchange passengers. This paper introduces pedestrian detection into the statistics to realize real-time statistics of passenger transfer flow. This article first determines the research scenario and simplifies the transfer path. Then, the pedestrian detection model is constructed by using the histogram of gradient directions (HOG) and support vector machine (SVM), and combined with the continuous adaptive mean shift algorithm (CamShift) to design and realize the transfer path passenger flow statistics. Finally, the algorithm was trained and tested through the samples obtained at Xi’an Xiaozhai subway station, and the test results obtained have high accuracy.

Human Interaction Behavior Recognition based on Local Spatial - Temporal and Global Feature

At present, the human interaction behavior algorithm has received great attention, but how to effectively extract the interactive characteristics of human interaction and establish the complex model between multiple targets is still a difficult problem. Aiming at this problem, we proposed a human interaction recognition method based on local spatial - temporal and global feature. This method treats people who are interacting as a whole. Firstly, foreground human object is extracted by the inter-frame difference. Secondly, the optical flow and space-time points of interest are extracted. Considering the large and high dimension of the feature, the characteristics of each frame image are described by the gradient direction histogram HOG and the optical flow histogram HOF. Then, BP neural network is used to classify each feature of each frame. Considering the limitations of different features for interactive behavior recognition, BP neural network combined with classification fusion is used to obtain the final interaction behavior category. The algorithm tested in the UT interaction dataset, and the experimental results show that this method can recognize the interaction behaviors of hitting, hugging, kicking, pushing and handshaking. The proposed method can be applied to intelligent video surveillance.

An improved local binary pattern method for pollen image classification and recognition

Most of the current feature extraction algorithms in pollen images perform recognition by extracting the contour, shape, texture, and spatial and frequency domain features of the divided images. For improving the effect of recognition, the improved Local Binary Pattern method is proposed for pollen image classification and recognition. Firstly, the image features are described by calculating and counting the directional histogram of the image's gradient. Secondly, the obtained image normalization processing result determines and generates corresponding samples, and multi-scale features are performed on the pollen images. Finally, the Histogram of Oriented Gradients are fused and binary coded by Local Binary Pattern method, and the similarity of each image is been calculated using Manhattan Distance. The experimental results use the Correct Recognition Rate and the Average Recognition Time. The experimental results can show that the proposed method is superior to other methods and robust to noise and rotation of pollen images.

The Role of 5G Network Image Information Based on Deep Learning in UAV Prediction Target Trajectory Tracking

With the development of information technology in the network era and the popularization of the 5G era, UAV‐related applications are becoming more and more widely used, which is one of the essential basic technologies. Therefore, the technology has great research value and practical significance, a multiobjective detector based on support vector machine (SVM) is designed based on directional gradient histogram (HOG), and the startup method used with cross‐validation methods can improve detector performance. It makes the detector accuracy above 98% and has good resistance to the target scale. A real‐time target tracker is designed with its rotation variation and with an improved average displacement algorithm. The algorithm must manually select the target model and suggest the target model to achieve automatic acquisition of the target model. Due to the ambiguity of the target tracking state, several judgment conditions are set to determine whether the tracking has failed and whether the tracker state is correctly verified, with several similar target tracking algorithms. When the system is started, the system detects targets frame by frame. And it will locate a possible target by color segmentation and specify the target to be tracked to recommend the relevant model during the tracking process and open the tracker to determine the target tracking state frame by frame and perform target detection at each frame. Then it will find possible goals and will follow them to achieve a balance of stable and real‐time system performance, using the results of the TPD‐KCF method. The percentage of correctly tracking images can reach 98%, and the efficiency is significantly improved.

Fast Rotation Matching Method for SAR and Optical Images

The rotation matching between synthetic aperture radar (SAR) and optical images is essential for their registration. In this article, we introduce a novel fast rotation matching method for SAR and optical images. In the method, the absolute directional derivative histograms (ADDHs) of SAR and optical images can be used to reflect the rotation difference between them, and then the rotation difference is removed by the cross correlation of the two images’ ADDHs. Subsequently, we generate the ADDHs by oriented Gaussian-Gamma-shaped operator. Then, the cross correlation based on fast Fourier transform is used to deal with the ADDHs of SAR and optical images in order to determine the rotation difference of the two images. The experiment results demonstrate that the proposed method can detect the rotation difference in various conditions and achieves rotation matching with high speed.

Multi-scale Correlation Filter Tracking Algorithm Based on Feature Fusion

For the Kernel Correlation Filter (KCF) algorithm using a single feature can not fully describe the tracking target, and is susceptible to the change of the target scale, lacking the estimation of the target scale size, an improved feature fusion and multi-scale correlation filter tracking algorithm based on KCF is proposed. Firstly, in the stage of location prediction, color features and directional gradient histogram hog features are fused by feature series, and feature dimensions are added to realize target location prediction. Then the scale filter is introduced into the predicted target position to estimate the optimal target scale as the tracking result. Experimental results show that the accuracy and success rate of the improved algorithm are significantly improved compared with other classical tracking algorithms, and it can deal with scale change, occlusion, deformation and other complex situations robustly.

Research on Target Tracking Algorithm Based on context Multi-feature Fusion

In traditional target tracking algorithms, cosine window is often used to suppress the boundary effect and expand the search range when the target is located, which will reduce the available background information. Besides, when a single feature is used, the representation of the target appearance is not strong. When the challenge factors such as background clutter and occlusion occur to the target, the target will drift, which will reduce the accuracy of target tracking and even lead to tracking failure. To solve this problem, this paper proposes a context-based target tracking algorithm that fuses multiple features, using directional gradient histogram features and color name features as input features. To solve the occlusion problem, the occlusion mechanism of average peak correlation energy and maximum response fraction value is designed to judge the updating of the model. When both values are higher than a certain historical mean, the model is updated; Otherwise, stop updating. Finally, experiments show that the tracking algorithm proposed in this paper is more robust than some traditional correlation filtering algorithms.

Segmentation of Lath-Like Structures via Localized Identification of Directionality in a Complex-Phase Steel

In this work, a segmentation approach based on analyzing local orientations and directions in an image, in order to distinguish lath-like from granular structures, is presented. It is based on common image processing operations. A window of appropriate size slides over the image, and the gradient direction and its magnitude inside this window are determined for each pixel. The histogram of all possible directions yields the main direction and its directionality. These two parameters enable the extraction of window positions which represent lath-like structures, and procedures to join these positions are developed. The usability of this approach is demonstrated by distinguishing lath-like bainite from granular bainite in so-called complex-phase steels, a segmentation task for which automated procedures are not yet reported. The segmentation results are in accordance with the regions recognized by human experts. The approach’s main advantages are its use on small sets of images, the easy access to the segmentation process and therefore a targeted adjustment of parameters to achieve the best possible segmentation result. Thus, it is distinct from segmentation using deep learning which is becoming more and more popular and is a promising solution for complex segmentation tasks, but requires large image sets for training and is difficult to interpret.

Human motion recognition based on limit learning machine

With the development of technology, human motion capture data have been widely used in the fields of human–computer interaction, interactive entertainment, education, and medical treatment. As a problem in the field of computer vision, human motion recognition has become a key technology in somatosensory games, security protection, and multimedia information retrieval. Therefore, it is important to improve the recognition rate of human motion. Based on the above background, the purpose of this article is human motion recognition based on extreme learning machine. Based on the existing action feature descriptors, this article makes improvements to features and classifiers and performs experiments on the Microsoft model specific register (MSR)-Action3D data set and the Bonn University high density metal (HDM05) motion capture data set. Based on displacement covariance descriptor and direction histogram descriptor, this article described both combine to produce a new combination; the description can statically reflect the joint position relevant information and at the same time, the change information dynamically reflects the joint position, uses the extreme learning machine for classification, and gets better recognition result. The experimental results show that the combined descriptor and extreme learning machine recognition rate on these two data sets is significantly improved by about 3% compared with the existing methods.

Enhanced Forest Microexpression Recognition Based on Optical Flow Direction Histogram and Deep Multiview Network

In order to recognize the instantaneous changes of facial microexpressions in natural environment, a method based on optical flow direction histogram and depth multiview network to enhance forest microexpression recognition was proposed. In the preprocessing stage, the histogram equalization of the acquired face image is performed, and then the dense key points of the face are detected. According to the coordinates of the key points and the face action coding system (FACS), the face region is divided into 15 regions of interest (ROI). In the feature extraction stage, the optical flow direction histogram feature between adjacent frames in ROI is extracted to detect the peak frame of microexpression sequence. Finally, the average optical flow direction histogram feature of the image sequence from the initial frame to the peak frame is extracted. In the classification stage, firstly, the head pose parameters under horizontal degrees of freedom are estimated to eliminate the influence of head pose motion, and a forest multiview conditional probability model based on deep multiview network is established. Conditional probability and neural connection function are introduced into the node splitting learning of random tree to improve the learning ability and distinguishing ability of the model on the limited training set. Finally, multiview-weighted voting is used to determine the categories of facial microexpressions. Experiments on CASME II microexpression dataset show that the proposed method can effectively describe the changes of microexpressions and improve the recognition accuracy compared with other new methods.

Writer identification with n ‐tuple direction feature from contour

This study introduces an effective solution for text-independent writer identification by generalising contour-hinge feature, which is called n-tuple direction feature. For extracting n-tuple direction feature, the authors first obtain all contours from connected components, then n + 1 points are considered on the contour with a certain distance apart, and next, the directions of the fragments connecting two successive points are computed. The n + 1 points move on the contour and the n-dimensional histogram of directions is computed. The proposed method is evaluated on large Farsi and English databases. A correct writer identification rate of 92.2% for English handwritings from 900 persons and 97.7% for Farsi handwritings from 600 persons are achieved. Comparison between the proposed method and other studies shows the promising performance and superiority of the proposed method.

Homomorphic Encryption-Based Robust Reversible Watermarking for 3D Model

Robust reversible watermarking in an encrypted domain is a technique that preserves privacy and protects copyright for multimedia transmission in the cloud. In general, most models of buildings and medical organs are constructed by three-dimensional (3D) models. A 3D model shared through the internet can be easily modified by an unauthorized user, and in order to protect the security of 3D models, a robust reversible 3D models watermarking method based on homomorphic encryption is necessary. In the proposed method, a 3D model is divided into non-overlapping patches, and the vertex in each patch is encrypted by using the Paillier cryptosystem. On the cloud side, in order to utilize addition and multiplication homomorphism of the Paillier cryptosystem, three direction values of each patch are computed for constructing the corresponding histogram, which is shifted to embed watermark. For obtaining watermarking robustness, the robust interval is designed in the process of histogram shifting. The watermark can be extracted from the symmetrical direction histogram, and the original encrypted model can be restored by histogram shifting. Moreover, the process of watermark embedding and extraction are symmetric. Experimental results show that compared with the existing watermarking methods in encrypted 3D models, the quality of the decrypted model is improved. Moreover, the proposed method is robust to common attacks, such as translation, scaling, and Gaussian noise.

Research on Dynamic Measurement Method of Electromagnetic Environment Based on Image Similarity

In this paper, by discretizing the battlefield region, using the power, frequency and direction of the wave to quantitatively characterize the electromagnetic environment, the image similarity measure, the gray histogram comparison and the signal arrival direction histogram method are introduced to analyze the histogram. The degree of change of the graph is proposed, and a calculation method for calculating the degree of change of electromagnetic environment is proposed. The method is simulated. The comparison of the histograms at different time points proves that the method has certain effectiveness in dealing with various electromagnetic environments.

Improving the Accuracy for Offline Arabic Digit Recognition Using Sliding Window Approach

Handwritten Digit recognition is a challenging problem these days due to the widely used Arabic language in the world, especially in the Middle East region. In this paper, sliding windows are used to enhance classification accuracies and implemented using random forests (RF) and support vector machine (SVM) classifiers for recognition of Arabic digit images. In order to study their effectiveness with and without using sliding windows, four different feature extraction techniques have been proposed which includes Mean-based, Gray-Level Co-occurrence Matrix (GLCM), Moment-based, and Edge Direction Histogram (EDH). The obtained accuracies show the significance of using sliding windows for classifying digit. The recognition rates acquired using the modified version of AHDBase dataset are 98% when Mean-based and Moment-based are applied with RF classifier, 98.33% and 99.13% when GLCM and EDH are used with linear-kernel SVM, respectively. Moreover, the performance of this study is compared against recent state-of-the-art approaches, namely Geometric-based, two-dimensional discrete cosine transform, Hierarchical features, Hetero-features, Discrete Fourier Transform and geometrical features, Gabor-based, gradient, structural, and concavity and Local Binary Convolutional Neural Networks.

A Novel Approach for Pose Invariant Face Recognition in Surveillance Videos

The face detection and face recognition methods are introduced to confirm the abnormal human activity in the video surveillance system. Face detection is carried out by Viola-Jones face detector. It is composed of three concepts namely integral image, AdaBoost and the cascade structure. After face detection, Histogram of Oriented Gradient(HOG) and Weighted Local Binary Patterns(WLBP) features are extracted and those are used in Orthogonal Locality Preserving Projection(OLPP) for face recognition. The detected faces may contain pose variation which dramatically degrades the OLPP based face recognition. So, pose-invariant OLLP-based face recognition is proposed where Histogram of Face Orientation(HFO) and Histogram of Face Direction(HFD), HOG and WLBP features are used in OLPP for efficient face recognition.

A Multi-Information Fusion Correlation Filters Tracker

In recent years, trackers based on correlation filters have attracted more and more attention due to the impressive tracking accuracy and real-time performance. However, in real scenarios, the tracking results are often been interfered with by the occlusion, illumination variation, appearance variation and background clutter. In order to find a tracker with better tracking performances, this paper proposed a multi-information fusion correlation filter tracker, which uses channel and spatial reliabilities and time regularization information on samples for filter training, and which not only extends the target search areas but also has a stronger ability to track the targets with significant appearance variations. Thus, results from extensive experiments conducted on OTB100, VOT2016, TC128, and UAV123 data sets show that our tracker with only directional gradient histogram (HOG) and color name (CN) features, performs favorably against the state-of-the-art trackers in terms of tracking precision, tracking success rate, tracking accuracy, and A-R rank.

Multi-level colored directional motif histograms for content-based image retrieval

Color features and local geometrical structures are the two basic image features which are sufficient to convey the image semantics. Both of these features show diverse nature on the different regions of a natural image. Traditional local motif patterns are standard tools to emphasize these local visual image features. These motif-based schemes consider either structural orientations or limited directional patterns which are not sufficient to realize the detailed local geometrical properties of an image. To address these issues, we have proposed a new multi-level colored directional motif histogram (MLCDMH) for devising a content-based image retrieval scheme. The proposed scheme extracts local structural features at three different levels. Initially, MLCDMH scheme extracts directional structural patterns from a $$3 \times 3$$ pixel grids of an image. This reflects the $$9^9$$ different structural arrangements using 28 directional patterns. Next, we have used a weighted neighboring similarity (WNS) scheme to exploit the uniqueness of each motif pixel in its local surrounding. The WNS scheme will compute the importance of each directional motif pattern in its $$3 \times 3$$ local neighborhood. In the last level, we have fused all directional motif images into a single directional difference matrix which reflects the local structural and directional motif features in detail and also reduces the computation overhead. The MLCDMH considers all possible permutations and rotations of the motif patterns to generate rotational invariant structural features. The image retrieval performance of this proposed scheme has been evaluated using different Corel/natural, object, texture and heterogeneous image datasets. The results of the retrieval experiments have shown satisfactory improvement over other motif- and non-motif-based CBIR approaches.

Crater matching algorithm based on feature descriptor

In the field of visual navigation, crater is an ideal navigation landmark on planet surface, because of its universality and significance. During the descent phase of lander, crater needs to be extracted and matched. However, as the landing proceeds, images taken by the on-board camera will change, such as the variation of scale, translation, rotation, illumination which will increase the difficulty of the matching algorithm.This paper proposes a more accurate and faster crater matching algorithm based on feature descriptor. This algorithm is designed for navigation during the descending phase. It has the invariance of scaling, rotation, illumination. First, this algorithm extracts circular arc by Gaussian Pyramid and ELSD algorithm, so that this algorithm has the invariance against scale variation. Then, in order to describe the circular arc, this algorithm determine the direction of the circular arc by the direction histogram. And this algorithm constructs the circular arc band descriptors in the support region to achieve translation, rotation and illumination invariance. At last, matching criteria is the combination of the Nearest Neighbor Distance Ratio (NNDR) and the Euclidean distance constraint. The results are obtained under different image variations. The matching results show that this crater matching algorithm has high crater correct extracting and matching rate and high computationally efficient.