Sequence Of Strokes Research Articles

Simulation of Tennis Behaviour Using Finite Markov Chains

Markov chains are a well-known tool in mathematical modelling. In performance analysis, finite Markov chains and the connected transition probabilities can be used to evaluate game behavior. Herein finite Markov chains are especially suited in the modelling of net games like tennis, due to their structure as an alternating sequence of discrete strokes. Finite Markov chain modelling thereby can be used as a descriptive tool, as well as to gain insight in the relationship between sports behavior and outcomes. The transition matrix can be used to display the frequency of game actions in relation to the respective game structure. Likewise, the model calculations from the theory of Markov chains enable determining the relevance of actions therein, concerning overall performance. To permit the usage of finite Markov chain modelling, adherence to the Markov property, often described as the property of “memorylessness”, must be assured. Present study implemented finite Markov chain modelling for elite level tennis at the Australian and French Open. The goal of the present study was to verify the usage of Markov chain modelling in tennis, using a newly designed transition matrix. Furthermore, the aim was to gain insight in the game structure of tennis. Results showed that the new model adhered to the Markov property in an extent, which allowed the usage of selected model predictions for the analysis. Further the analysis revealed insights in the current game structure of tennis, as well as connected influence of the factors court surface and gender of players.

TIVEE: Visual Exploration and Explanation of Badminton Tactics in Immersive Visualizations.

Tactic analysis is a major issue in badminton as the effective usage of tactics is the key to win. The tactic in badminton is defined as a sequence of consecutive strokes. Most existing methods use statistical models to find sequential patterns of strokes and apply 2D visualizations such as glyphs and statistical charts to explore and analyze the discovered patterns. However, in badminton, spatial information like the shuttle trajectory, which is inherently 3D, is the core of a tactic. The lack of sufficient spatial awareness in 2D visualizations largely limited the tactic analysis of badminton. In this work, we collaborate with domain experts to study the tactic analysis of badminton in a 3D environment and propose an immersive visual analytics system, TIVEE, to assist users in exploring and explaining badminton tactics from multi-levels. Users can first explore various tactics from the third-person perspective using an unfolded visual presentation of stroke sequences. By selecting a tactic of interest, users can turn to the first-person perspective to perceive the detailed kinematic characteristics and explain its effects on the game result. The effectiveness and usefulness of TIVEE are demonstrated by case studies and an expert interview.

Auditing an insightful method for determining the sequence of intersecting strokes: Scanning electron microscope

Over the years, many studies have been published on complex and challenging problem of sequence of strokes examination in order to find a suitable technique in the field of questioned documents. There are various destructive and non-destructive techniques for the examination of various sequence problems. Scanning electron microscopy (SEM) or scanning electron microscopy with energy dispersive x-ray (SEM-EDX) could be one of the advanced and complementary technique useful in chronological investigation of sequence of strokes. This paper deals with the brief introduction to the scanning electron microscope along with research studies conducted on determining sequence of intersecting strokes utilizing SEM.

Determination of Sequence of Strokes-Conventional & Non-Conventional Techniques - A Review Study

The most important tool for a document examiner is his or her own vision, but in this section and the laboratory tour section we will consider the external tools which enable the examiner to see, evaluates and record more than is apparent to the unaided eye. The number of papers outlines the basic necessity of photographic illustrations for demonstrating the data offer clinching evidence, strong enough to with stand legal scrutiny on which opinions of the document examiner are based, hence play a very vital role in the examination of forensic documents. For the purpose of examination in the laboratory as well as the use of photographs becomes indispensible in cases involving traced or simulated forgery, built-up documents, forgery over genuine signatures and forgery by substitution, interpolations, differences in the tint of ink etc. The authors have solved a case in resolving the sequence of strokes with same ball point ink in an actual case received in the laboratory.

RallyComparator: visual comparison of the multivariate and spatial stroke sequence in table tennis rally

RallyComparator: visual comparison of the multivariate and spatial stroke sequence in table tennis rally

Towards Position-Independent Sensing for Gesture Recognition with Wi-Fi

Past decades have witnessed the extension of the Wi-Fi signals as a useful tool sensing human activities. One common assumption behind it is that there is a one-to-one mapping between human activities and Wi-Fi received signal patterns. However, this assumption does not hold when the user conducts activities in different locations and orientations. Actually, the received signal patterns of the same activity would become inconsistent when the relative location and orientation of the user with respect to transceivers change, leading to unstable sensing performance. This problem is known as the position-dependent problem, hindering the actual deployment of Wi-Fi-based sensing applications. In this paper, to tackle this fundamental problem, we develop a new position-independent sensing strategy and use gesture recognition as an application example to demonstrate its effectiveness. The key idea is to shift our observation from the traditional transceiver view to the hand-oriented view, and extract features that are irrespective of position-specific factors. Following the strategy, we design a position-independent feature, denoted as Motion Navigation Primitive(MNP). MNP captures the pattern of moving direction changes of the hand, which shares consistent patterns when the user performs the same gesture with different position-specific factors. By analyzing the pattern of MNP, we convert gestures into sequences of strokes (e.g, line, arc and corner) which makes them easy to be recognized. We build a prototype WiFi gesture recognition system, i.e., WiGesture to validate the effectiveness of the proposed strategy. Experiments show that our system can outperform the start-of-arts significantly in different settings. Given its novelty and superiority, we believe the proposed method symbolizes a major step towards gesture recognition and would inspire other solutions to position-independent activity recognition in the future.

Leader-chasing behavior in negative artificial triggered lightning flashes

Two special cases of dart leader propagation were observed by the high-speed camera in the leader/return stroke sequences of a classical triggered lightning flash and an altitude-triggered lightning flash, respectively. Different from most of the subsequent return strokes preceded by only one leader, the return stroke in each case was preceded by two leaders occurring successively and competing in the same channel, which herein is named leader-chasing behavior. In one case, the polarity of the latter leader was opposite to that of the former leader and these two combined together to form a new leader, which shared the same polarity with the former leader. In the other case, the latter leader shared the same polarity with the former leader and disappeared after catching up with the former leader. The propagation of the former leader in this case seems not to be significantly influenced by the existence of the latter leader.

Content Masked Loss: Human-Like Brush Stroke Planning in a Reinforcement Learning Painting Agent

The objective of most Reinforcement Learning painting agents is to minimize the loss between a target image and the paint canvas. Human painter artistry emphasizes important features of the target image rather than simply reproducing it. Using adversarial or L2 losses in the RL painting models, although its final output is generally a work of finesse, produces a stroke sequence that is vastly different from that which a human would produce since the model does not have knowledge about the abstract features in the target image. In order to increase the human-like planning of the model without the use of expensive human data, we introduce a new loss function for use with the model's reward function: Content Masked Loss. In the context of robot painting, Content Masked Loss employs an object detection model to extract features which are used to assign higher weight to regions of the canvas that a human would find important for recognizing content. The results, based on 332 human evaluators, show that the digital paintings produced by our Content Masked model show detectable subject matter earlier in the stroke sequence than existing methods without compromising on the quality of the final painting. Our code is available at https://github.com/pschaldenbrand/ContentMaskedLoss.

An upward negative lightning flash triggered by a distant +CG from a tall tower in Florida: Observations and modeling

We examined in detail the morphology and evolution of an upward negative flash initiated from a 257-m tower in Florida. High-speed video camera images and wideband electric field records, as well as ENTLN data, were used. The upward negative flash was induced (triggered) by a single-stroke 50-kA +CG that occurred about 45 km from the tower. The 257-m tower flash contained 6 leader/return stroke sequences and 1 attempted leader that almost terminated on the tower. All the leaders exhibited bidirectional extension. Electric field waveforms produced by the return strokes (measured at a distance of 8.8 km) were bipolar and abnormally narrow (exhibited earlier zero crossings). In order to examine the origin of the observed earlier zero crossings, we used two return-stroke models of transmission line type (MTLL and MTLE) to see what model input parameters are responsible for this feature. Within the limits of those models, the observed earlier zero crossings could be explained only by a narrow input current waveform or/and its fast amplitude decay with height.

Determination of crossing sequences between seal impression and laser-printed lines based on characteristics of the toner molten state.

Determining the sequence of intersection lines is a challenge in forensic document examination. To explore more advanced technology and meet the needs of practical applications, this study proposes a method based on the characteristics of the molten state to examine crossing-line sequences. This state can be considered as a melting state of solid laser toner through a dissolution process. Experimental results demonstrate that the molten state is always visible at intersections when a particular ink is stamped after laser printing; however, the presence of the molten state when the seal is stamped before laser printing largely depends on the time interval between printing and stamping. The degree of error obtained through blind tests indicates that this approach has its limitations; however, a high accuracy, which refers to the ability of an examiner, was achieved with respect to determining the sequence of strokes when using a fluorescence method in conjunction with this method. This further indicated the limitation of a single method and the importance of combining multiple methods for document examinations.

DCR: Disentangled component representation for sketch generation

• We propose a novel end-to-end model to extract the disentangled representation of a sketch. • We discuss the performance of existing methods on the multi-class sketch generation task. • We investigate why disentangled representation can improve the performance of sketch generation. • We release a large online handwritten Chinese character dataset with component-level and image-level labels. Our model consists of three modules, a CNN autoencoder for extracting image-level representations f conv , an LSTM autoencoder for extracting component representations z t , and a feature disentangling module for disentangling the image-level representations to the component representations. Compared to directly mapping the pixel image into the whole sketch sequences, our model explores disentangling representations and gets a further understanding of components shared among sketches to make sketch generation more interpretable, flexible and diversifiable. We present a simple end-to-end model based on deep learning to automatically decompose sketched objects into components by disentangling the visual representation. The performance of visual representation learning based models degrades as categories increase. Rather than building a mapping from a static image to the whole sketch sequences, we propose an interpretable disentangled representation of sketch to understand component concepts and the relationship among such concepts. Our model takes the binary image of a sketched object and produces a component stroke sequence set corresponding to key components in the sketch. Experiments show that our method significantly outperforms all baselines quantitatively at the degree of disentanglement, and our method is more stable while training on tens of categories.

Research on Recognition Effect of DSCN Network Structure in Hand-Drawn Sketch.

With the rapid development of image recognition technology, freehand sketch recognition has attracted more and more attention. How to achieve good recognition effect in the absence of color and texture information is the key to the development of freehand sketch recognition. Traditional nonlearning classical models are highly dependent on manual selection features. To solve this problem, a neural network sketch recognition method based on DSCN structure is proposed in this paper. Firstly, the stroke sequence of the sketch is drawn; then, the feature is extracted according to the stroke sequence combined with neural network, and the extracted image features are used as the input of the model to construct the time relationship between different image features. Through the control experiment on TU-Berlin dataset, the results show that, compared with the traditional nonlearning methods, HOG-SVM, SIFT-Fisher Vector, MKL-SVM, and FV-SP, the recognition accuracy of DSCN network is improved by 15.8%, 10.3%, 6.0%, and 2.9%, respectively. Compared with the classical deep learning model, Alex-Net, the recognition accuracy is improved by 5.6%. The above results show that the DSCN network proposed in this paper has strong ability of feature extraction and nonlinear expression and can effectively improve the recognition accuracy of hand-painted sketches after introducing the stroke order.

SketchFormer: transformer-based approach for sketch recognition using vector images

Sketches have been employed since the ancient era of cave paintings for simple illustrations to represent real-world entities and communication. The abstract nature and varied artistic styling make automatic recognition of these drawings more challenging than other areas of image classification. Moreover, the representation of sketches as a sequence of strokes instead of raster images introduces them at the correct abstract level. However, dealing with images as a sequence of small information makes it challenging. In this paper, we propose a Transformer-based network, dubbed as AttentiveNet, for sketch recognition. This architecture incorporates ordinal information to perform the classification task in real-time through vector images. We employ the proposed model to isolate the discriminating strokes of each doodle using the attention mechanism of Transformers and perform an in-depth qualitative analysis of the isolated strokes for classification of the sketch. Experimental evaluation validates that the proposed network performs favorably against state-of-the-art techniques.

Attention-Net: An Ensemble Sketch Recognition Approach Using Vector Images

For the past few decades, machines have replaced humans in several disciplines. However, machine cognition still lags behind the human capabilities. We address the machines’ ability to recognize human drawn sketches in this work. Visual representations, such as sketches have long been a medium of communication for humans. For artificially intelligent systems to effectively immerse in interactive environments, it is required that machines understand such notations. The abstract nature and varied artistic styling of these sketches make automatic recognition of drawings more challenging than other areas of image classification. In this article, we use sketches represented as a sequence of strokes, i.e., as vector images, to effectively capture the long-term temporal dependencies in hand-drawn sketches. The proposed approach combines the self-attention capabilities of Transformers while effectively utilizing the long-term temporal dependencies through temporal convolution networks (TCNs) for sketch recognition. The confidence scores obtained from the two techniques are combined using triangular-norm (T-norm). Attention heat maps are plotted to isolate the discriminating parts of a sketch that contribute to sketch classification. The extensive quantitative and qualitative evaluation confirms that the proposed network performs favorably against state-of-the-art techniques.

Towards a novel biometric system for forensic document examination

Biometric systems have demonstrated an important enhancement in writer identification from online handwriting. Indeed, writer identification is still a challenging task in the definition of a set of features able to characterize the different samples of handwriting documents. These samples are not usually stable and show a vast variability from the same writer over time, or from different writers. The ability to identify the documents’ authors provides more chances for using these documents for various purposes and especially in forensic analysis. In this paper, we present a biometric based recognition system for forensic document examination capable of identifying a document's author. This system consists of the preprocessing and the segmentation of online handwriting into a sequence of strokes in a first step. Then, from each stroke, we extract a set of static and dynamic features. Next, all the segments which are composed of two consecutive strokes are categorized into groups and subgroups according to their position and their geometric characteristics. Finally, Deep Neural Network is used as a classifier. Experiments which are conducted on IBM_UB_1, IAM OnDB, NLPR Handwriting and ADAB databases, demonstrate that the proposed system outperforms the existing writer identification systems on Latin and Arabic scripts. This justifies its usefulness in forensic examination of handwriting.

Determination of the Chronological Order of a Laser Printing and Gel Pen Writing Without any Crossing Strokes

Laser printers and writing instruments are being exploited by the forgers to produce fabricated documents. They often produce such documents by appending genuine signature on a blank sheet and later printing on it. Often these fabricated documents do not contain intersection of strokes and may pose a difficulty to forensic document examiners in analyzing cases related to such documents. Keeping in mind the limited research work done on nonintersecting strokes, the present study has been conducted with an aim to determine the sequence of nonintersecting strokes of gel pens and laser printers. Three hundred samples of nonintersecting strokes of gel pen ink and laser printing were prepared. Of these 300, 150 had laser-printed text below the gel pen strokes and the remaining 150 were prepared with gel pen stokes below the laser-printed text. The samples have been analyzed using Nikon SMZ 800N stereomicroscope. Three micro-conformation features, that is, pattern of toner distribution, ink distribution, and fiber distribution have been observed in the samples. The results demonstrate the successful establishment of sequence of nonintersecting strokes in case of gel pen and laser-printed documents by the determination of micro-conformation features.

Observations of the Wire Destruction and Plasma Channel Reestablishment Process During the Initial Stage of Triggered Lightning

AbstractThe wire destruction and plasma channel reestablishment process during the initial stage of a triggered lightning is analyzed based on comprehensive observations, including very high frequency (VHF) mapping array, electric‐magnetic field, and high‐speed video. The detailed discharge processes in both the attempted reconnection process and reconnection process are firstly evidenced by VHF mapping results. The reconnection process is similar to a leader/return‐stroke sequence, while the attempted reconnection process involves more complicated discharges. It is shown that several segments of the wire are fused firstly due to inhomogeneity of the wire, resulting in a series of wire residues. Thereafter, downward extending VHF breakdowns dispersed over hundreds of meters in altitude followed by a recoil‐like event, both of which emitted fairly weak VHF radiation, and no detectable LF electromagnetic field change was observed. It is indicated that there are no significant current flowing or charges movement during the attempted reconnection process.

Deep imitator: Handwriting calligraphy imitation via deep attention networks

Calligraphy imitation (CI) from a handful of target handwriting samples is such a challenging task that most of the existing writing style analysis or handwriting generation methods do not exhibit satisfactory performance. In this paper, we propose a novel multi-module framework to address the problem of CI. Firstly, we utilized a deep convolution neural network (CNN) to extract personalized calligraphical features. Then we built a calligraphy-clustering attention module and a mata-style matrix (msM) to compute an embedding of calligraphy. The structure of conditional gated recurrent unit (cGRU) is then improved to predict the probabilistic density of pen tip movement displacement by dual condition inputs. Finally, we generated personalized handwriting stroke sequences through iterative sampling with Gaussian mixture model (GMM). Experiments on public online handwriting databases verify that the proposed method could achieve satisfactory performance; the generated samples achieved high similarities with original handwriting examples.

Stroke Sequence-Dependent Deep Convolutional Neural Network for Online Handwritten Chinese Character Recognition.

We propose a novel model, called stroke sequence-dependent deep convolutional neural network (SSDCNN), which uses the stroke sequence information and eight-directional features of Chinese characters for online handwritten Chinese character recognition (OLHCCR). SSDCNN learns the representation of OLHCCs by incorporating the natural sequence information of the strokes. Furthermore, it naturally incorporates the eight-directional features. First, SSDCNN inputs the stroke sequence and transforms it into stacks of feature maps following the writing order of the strokes. Second, the fixed-length, stroke sequence-dependent representations of OLHCC are derived through convolutional, residual, and max-pooling operations. Third, the stroke sequence-dependent representation is combined with the eight-directional features via a number of fully connected neural network layers. Finally, the Chinese characters are recognized using a softmax classifier. The SSDCNN is trained in two stages: 1) the whole architecture is pretrained using the training data until the performance converges to an acceptable degree. 2) The stroke sequence-dependent representation is combined with the eight-directional features by a fully connected neural network and a softmax layer for further training. The model was experimentally evaluated on the OLHCCR competition tasks of International Conference on Document Analysis and Recognition (ICDAR) 2013. The recognition error was a maximum 58.28% lower in SSDCNN than in a model using the eight-directional features alone (5.13% versus 2.14%). Owing to its high accuracy (97.86%), the proposed SSDCNN reduced the recognition error by approximately 18.0% as compared with that of the winning system in the ICDAR 2013 competition. SSDCNN integrated with an adaptation mechanism, called the SSDCNN+Adapt model, and reached a new state-of-the-art (SOTA) standard with an accuracy of 97.94%. The SSDCNN exploits the stroke sequence information to learn high-quality OLHCC representations. Moreover, the learned representation and the classical eight-directional features complement each other within the SSDCNN architecture.

Establishing Sequence of Inkjet Printer, Laser Printer and Writing Ink Strokes using Scanning Electron Microscopy (SEM)

The aim of the present study is to determine the sequence of laser printer, inkjet printer, and writing ink (ball point pen ink, gel pen ink, pilot pen ink and fountain pen ink) strokes using a low voltage (1kV) scanning electron microscope (SEM). Intersections were prepared using an inkjet printer, laser printer, and writing instruments on copier paper. About 1cm2 of intersection was cut using scissors and mounted to the probe stage of the SEM using double-sided tape. Each sample was analyzed at different operating parameters. The results were evaluated on the basis of continuity and iscontinuity of strokes at intersection point. It was possible to determine the correct sequence of printer strokes versus other writing instruments’ strokes; inkjet printer stroke versus fountain and pilot pen ink stroke; however, inconclusive results were obtained in the case of inkjet printer strokes versus ball point pen and gel pen ink strokes. The ffectiveness of this technique was determined by analyzing the exact sequence of blind samples. An SEM could be used as mplimentary tool with other optical methods to examine the sequence of strokes.