Input Character Research Articles

Text writing in the air

This paper presents a real-time video-based pointing method that allows the sketching and writing of English text over air in front of a mobile camera. The proposed method has two main tasks. First, it tracks the colored fingertip in the video frames and then applies English optical character reorganization over the plotted images to recognize the written characters. Moreover, the proposed method provides a natural human–system interaction without requiring a keypad, stylus, pen, glove, or any other device for character inputs. For the experiments, an application was developed using OpenCv with the JAVA language. The proposed method was tested on a Samsung Galaxy3 android mobile phone. The proposed algorithm showed an average accuracy rate of 92.083% when tested for different shaped alphabets. Here, more than 3000 different shaped characters were used. The proposed system is a software-based approach and is relevantly very simple, fast, and easy to use. It does not require sensors or any hardware other than a camera and red tape, and is applicable to all disconnected languages. It has one issue, though it is color-sensitive in that the existence of any red-colored object in the background before the start of and during the character writing can lead to false results.

License Plate Recognition Based on Neural Network Algorithm to Improve Research

with the development of intelligent transportation technology, which all countries are suitable for their own license plate recognition system is developed. But because of the CCD camera Angle problem will make license plate image tilt; Segmentation after do not match the characters in size and character discontinuity, led to license plate recognition rate is not high, speed slow, reduce the real-time performance of the system. In order to improve the rate of convergence, the recognition rate presents a license plate recognition algorithm based on BP neural network. First put the image correction, segmentation of character normalization processing and eliminate the unfavorable factors, then puts forward characteristics of characters input for training the BP neural network. By setting the network weights and training transfer function, improved algorithm to improve the recognition rate of the system, as well as the real-time performance.

Synthesizing Two-character Interactions by Merging Captured Interaction Samples with their Spacetime Relationships

Existing synthesis methods for closely interacting virtual characters relied on user-specified constraints such as the reaching positions and the distance between body parts. In this paper, we present a novel method for synthesizing new interacting motion by composing two existing interacting motion samples without the need to specify the constraints manually. Our method automatically detects the type of interactions contained in the inputs and determines a suitable timing for the interaction composition by analyzing the spacetime relationships of the input characters. To preserve the features of the inputs in the synthesized interaction, the two inputs will be aligned and normalized according to the relative distance and orientation of the characters from the inputs. With a linear optimization method, the output is the optimal solution to preserve the close interaction of two characters and the local details of individual character behavior. The output animations demonstrated that our method is able to create interactions of new styles that combine the characteristics of the original inputs.

Radical-stroke-based Chinese input method using dynamic radical keypad on touchscreen smartphones

This paper presents a new Chinese input method that utilises finger gestures for improving stroke-based input method on touchscreen devices. It is a radical-stroke (RS) based method with a dynamic radical keypad (DRK), in which single radical multiple strokes coding is used. It speeds up the input of complex Chinese characters by simply sliding a finger. Experimental results show that the proposed RS–DRK outperforms the conventional Chinese input methods on an Android platform.

Big-Screen Text Input System Based on Gesture Recognition

Now more and more based on gesture interaction system applications, but there are simple gestures to operate the mouse interaction, no text input to the system function. In this paper, the blind letters gestures as input gestures ,using kinect to get the depth image, gestures split, use SIFT feature extraction, then get manual alphabet as Pinyin input method to provide an Chinese character input to the system.

동일한 입력 문자를 가지는 상태의 병합을 통한 메모리 효율적인 결정적 유한 오토마타 구현

패턴 정합 알고리듬은 침입 탐지 및 방지 시스템의 성능을 좌우하는 중요한 기능 요소로서 일반적으로 정규 표현식(Regualr Expressions)을 사용해 패턴을 표현한다. 공격 패턴이 복잡해지고 다양해짐에 따라, 정규 표현식 또한 복잡해지고 그 수가 증가하고 있으며 이로 인해, 패턴 매칭 알고리듬에서 정규 표현식을 인식하기 위해 사용된 결정적 유한 오토마타(Deterministic Finite Automata)를 구성하는 상태가 폭발적으로 증가(states blowup)하고 있다. 이러한 상태의 폭발적 증가 문제를 해결하고 메모리 효율적인 자료 구조를 구현하기 위해 많은 연구가 이루어졌다. 대부분의 연구 결과들에서는 하나의 정규 표현식을 변환한 결정적 유한 오토마톤(Automaton) 내 상태의 수를 감소시키기 위한 효과적인 방안들을 제안하였다. 하지만, 이들 연구 결과는 단일 패턴 내 상태의 수만을 감소시킬 뿐 패턴의 수에 따라 증가하는 상태의 수를 감소시키지 못하는 한계점을 가지고 있다. 본 논문에서는 이를 해결하기 위해 정규 표현식으로 구성된 유한 오토마타(Automata) 상의 상태 병합을 통한 상태 감소 방안을 제안한다. 이는 동일한 입력 문자를 가지는 상태를 병합함으로써 유한 오토마타 상의 상태의 수를 감소시켜, 기존 결정적 유한 오토마타에 비해 평균 40.0%의 메모리 감소 효과를 나타낸다. A pattern matching algorithm plays an important role in traffic identification and classification based on predefined patterns for intrusion detection and prevention. As attacks become prevalent and complex, current patterns are written using regular expressions, called regexes, which are expressed into the deterministic finite automata(DFA) due to the guaranteed worst-case performance in pattern matching process. Currently, because of the increased complexity of regex patterns and their large number, memory-efficient DFA from states reduction have become the mainstay of pattern matching process. However, most of the previous works have focused on reducing only the number of states on a single automaton, and thus there still exists a state blowup problem under the large number of patterns. To solve the above problem, we propose a new state compression algorithm that merges states on multiple automata. We show that by merging states with the same input character on multiple automata, the proposed algorithm can lead to a significant reduction of the number of states in the original DFA by as much as 40.0% on average.

Devnagari Handwriting Recognition using STANN

approaches to recognize the handwritten characters and numerals such as online cursive handwriting and numerals recognition have been proposed. Most of them are based on neural network approaches. In this paper, a technique for continuous recognition of handwritten Devanagari characters is proposed. This approach employs a method based on Spatio-Temporal Artificial Neural Network (STANN). The proposed method is efficient in the field of online handwriting recognition because of the property of STANN to detection of spikes of the continuous input signals. The method is based on different steps of recognition. The results of signals are extracted from the input handwritten characters where the spikes signals are generated by the continuous signals of the input character. A new algorithm using STANN is proposed with results in this paper.

An Online Semi-supervised Brain–Computer Interface

Practical brain-computer interface (BCI) systems should require only low training effort for the user, and the algorithms used to classify the intent of the user should be computationally efficient. However, due to inter- and intra-subject variations in EEG signal, intermittent training/calibration is often unavoidable. In this paper, we present an online semi-supervised P300 BCI speller system. After a short initial training (around or less than 1 min in our experiments), the system is switched to a mode where the user can input characters through selective attention. In this mode, a self-training least squares support vector machine (LS-SVM) classifier is gradually enhanced in back end with the unlabeled EEG data collected online after every character input. In this way, the classifier is gradually enhanced. Even though the user may experience some errors in input at the beginning due to the small initial training dataset, the accuracy approaches that of fully supervised method in a few minutes. The algorithm based on LS-SVM and its sequential update has low computational complexity; thus, it is suitable for online applications. The effectiveness of the algorithm has been validated through data analysis on BCI Competition III dataset II (P300 speller BCI data). The performance of the online system was evaluated through experimental results on eight healthy subjects, where all of them achieved the spelling accuracy of 85 % or above within an average online semi-supervised learning time of around 3 min.

워키태깅 : 스마트폰 환경에서 음성기반의 효과적인 영상 콘텐츠 어노테이션 방법에 관한 연구

The rapid growth and dissemination of touch-based mobile devices such as smart phones and tablet PCs, gives numerous benefits to people using a variety of multimedia contents. Due to its portability, it enables users to watch a soccer game, search video from YouTube, and sometimes tag on contents on the road. However, the limited screen size of mobile devices and touch-based character input methods based on this, are still major problems of searching and tagging multimedia contents. In this paper, we propose WalkieTagging, which provides a much more intuitive way than that of previous one. Just like any other previous video tagging services, WalkieTagging, as a voice-based annotation service, supports inserting detailed annotation data including start time, duration, tags, with little effort of users. To evaluate our methods, we developed the Android-based WalkieTagging application and performed user study via a two-week. Through our experiments by a total of 46 people, we observed that experiment participator think our system is more convenient and useful than that of touch-based one. Consequently, we found out that voice-based annotation methods can provide users with much convenience and satisfaction than that of touch-based methods in the mobile environments.

Hybrid evaluation method for Korean Character input system in mobile phones

In this study, hybrid evaluation incorporating the Hick–Hyman law was proposed to predict input performance of a Korean input system applied to the number pads of mobile phones. The hybrid model, calculated based on the Hick–Hyman law, is a combination of reaction time (RT) and movement time (MT) model, which is a better model for predicting user performance than the conventional models that were used to assess Korean input system evaluation such as the Fitts' and the KLM-GOMS models. Along with Fitts' model (of visual-controlled movements), the simple movement time (SMT) and ballistic movement time (BMT) models (of ballistic behavior movement) were also evaluated to measure MT. Results showed that ballistic behavior movement was more compatible than visual-controlled movement. In ballistic behavior movement, a thumb-movement distance of <21 mm was closer to the SMT than the BMT model. These results indicate that when applying hybrid model movement, the displacement, and the target size properties of mobile phone users is required in the engineering of a predictive model. Thus, the suggested hybrid evaluation is meaningful in that it can evaluate users' input behaviors very appropriately in the case of hand-held input devices. Relevance to industryA hybrid model based on Hick–Hyman law was determined to be effective to evaluate text input systems for the hand-held input device. This effective hybrid model can provide more accurate prediction time for text input behaviors to its designers for efficient system design. To its usability engineers, this hybrid model for more accurate prediction time can also provide an effective quantitative tool for usability evaluations for hand-held input device manufacturing industries because this model can save efforts as well as time for usability testing.

Lexical and Parser tool for CBOOP program

This paper addresses an approach to build lexical and parser tool for Component Based Object Oriented Program (CBOOP). In this Paper,Lexical analysis tool use for the scanning of CBOOP program. Lexical tool reads the input characters of the source program and return the tokens. Parser will generate the syntax tree of CBOOP program and check the syntax of program. The driver program, i.e., main program will

A Study of EPUB 3.0 Authoring Tools Availability in Vertical Form Function

The purpose of this study was to identify the availability of authoring tools that support vertical flow function based upon empirical experiments. There were five EPUB authoring tools selected for this study. Four major editing procedures were designed for the project. The first was the support of Chinese character input. The second was vertical form editing. The third was ruby position editing, and finally, testing whether the editing software was capable of outputting a document as an e-Book in EPUB 3.0 format. Five programs were tested, and only one of the five was able to fulfill the aforementioned requirements. The usability of each program is positively verified through this study.

Most Probable Longest Common Subsequence for Recognition of Gesture Character Input

This paper presents a technique for trajectory classification with applications to dynamic free-air hand gesture recognition. Such gestures are unencumbered and drawn in free air. Our approach is an extension to the longest common subsequence (LCS) classification algorithm. A learning preprocessing stage is performed to create a probabilistic 2-D template for each gesture, which allows taking into account different trajectory distortions with different probabilities. The modified LCS, termed the most probable LCS (MPLCS), is developed to measure the similarity between the probabilistic template and the hand gesture sample. The final decision is based on the length and probability of the extracted subsequence. Validation tests using a cohort of gesture digits from video-based capture show that the approach is promising with a recognition rate of more than 98 % for video stream preisolated digits. The MPLCS algorithm can be integrated into a gesture recognition interface to facilitate gesture character input. This can greatly enhance the usability of such interfaces.

Neural Network-Based English Alphanumeric Character Recognition

Propose a neural-network based size and color invariant character recognition system using feed -forward neural network. Our feed-forward network has two layers. One is input layer and another is output layer. The whole recognition process is divided into four basic steps such as pre -processing, normalization, network establishment and recognition. Pre -processing involves digitization, noise remova l and boundary detection. After boundary detection, the input character matrix is normalized into 12×8 matrix for size invariant recognition and fed into the proposed network which consists of 96 input and 36 output neurons. Then we trained our network byproposed training algorithm in a supervised manner and established the network by adjusting weights. Finally, we have tested our network by more than 20 samples per character on average and give 99.99% accuracy only for numeric digits (0~9), 98% accuracy o nly for letters (A~Z) and more than 94% accuracy for alphanumeric characters by considering inter -class similarity measurement.

A Chinese Text Input Brain–Computer Interface Based on the P300 Speller

A visual speller is a brain–computer interface that empowers users with limited motor functionality to input text into a computer by measuring their electroencephalographic responses to visual stimuli. Most prior research on visual spellers has focused on input of alphabetic text. Adapting a speller for other types of segmental or syllabic script is straightforward because such scripts comprise sufficiently few characters that they may all be displayed to the user simultaneously. Logographic scripts, such as Chinese hanzi, however, impose a challenge: How should the thousands of Chinese characters be displayed to the user? Here, we present a visual speller, based on Farwell and Donchin's P300 Speller, for Chinese character input. The speller uses a novel shape-based method called the First–Last, or FLAST, method to encode more than 7,000 Chinese characters. Characters are input by selecting two components, from a set of 56 distinct components, that match the shape of the target character, followed by selection of the character itself. At the input speed of one character per 107 s, 24 able-bodied participants achieved mean online accuracy of 82.8% per component selection and 63.5% per character input. At the faster input speed of one character per 77 s, mean online accuracy was 59.4% per component selection and 33.3% per character input.

High-Performance and Compact Architecture for Regular Expression Matching on FPGA

We present the design, implementation and evaluation of a high-performance architecture for regular expression matching (REM) on field-programmable gate array (FPGA). Each regular expression (regex) is first parsed into a concise token list representation, then compiled to a modular nondeterministic finite automaton (RE-NFA) using a modified version of the McNaughton-Yamada algorithm. The RE-NFA can be mapped directly onto a compact register-transistor level (RTL) circuit. A number of optimizations are applied to improve the circuit performance: 1) spatial stacking is used to construct an REM circuit processing m ≥ 1 input characters per clock cycle; 2) single-character constrained repetitions are matched efficiently by parallel shift-register lookup tables; 3) complex character classes are matched by a BRAM-based classifier shared across regexes; 4) a multipipeline architecture is used to organize a large number of RE-NFAs into priority groups to limit the I/O size of the circuit. We implemented 2,630 unique PCRE regexes from Snort rules (February 2010) in the proposed REM architecture. Based on the place-and-route results from Xilinx ISE 11.1 targeting Virtex5 LX-220 FPGAs, the proposed REM architecture achieved up to 11 Gbps concurrent throughput for various regex sets and up to 2.67× the throughput efficiency of other state-of-the-art designs.

사용자 입력 문자에 반응하는 군집 로봇 행동 구현

This paper presents an approach to implement the behaviors of multi-robots responding to user's input characters. The robots are appropriately displaced to express any input characters. Using our method, any user can easily and friendly control multirobots. The responses of the robots to the user's input are intuitive. We utilize the centroidal Voronoi algorithm and the continuoustime Lloyd algorithm, which have popularly been used for the optimal sensing coverage problems. Collision protection is considered to be applied for real robots. LED sensors are used to identify positions of multi-robots. Our approach is evaluated through experiments with five mobile robots. When a user draw alphabets, the robots are deployed correspondingly. By checking position errors, the feasibility of our method is validated.

Dissolved Organic Matter Biogeochemistry Along a Transect of the Okavango Delta, Botswana

Biogeochemical processing of dissolved organic matter (DOM) in aquatic environments can alter its chemical quality and its bioavailability to the microbial loop. In this study, we evaluated the relative importance to DOM character of allochthonous and autochthonous DOM inputs and photo-degradation in a large, pristine wetland, the Okavango Delta of Botswana. We performed an intensive spatial sampling of surface water and analyzed for chemical and physical parameters (pH, conductivity, dissolved oxygen saturation, temperature, and channel depth), dissolved organic matter (DOM), and particulate organic matter (POM). We used UV–vis absorbance, fluorescence spectroscopy, and parallel factor analysis of excitation emission matrix data (EEM-PARAFAC) to characterize DOM. Our findings from principal component analysis (PCA) show downstream changes in DOM chemistry to be dominated by photo-degradation, suggesting that DOM in the Okavango Delta is transformed photo-chemically in shallower downstream reaches after being mobilized from the permanent swamp and seasonal floodplains. Additionally, we found that the PARAFAC model developed for the Everglades, a large, anthropogenically-altered wetland in North America, was well suited to tracking DOM dynamics in the Okavango Delta and may be useful for characterizing DOM in other sub-tropical, seasonally flooded wetlands.

Research of Off-Line Handwritten Chinese Character Recognition System Based on Feedback Structure

An offline handwritten Chinese character recognition system based on feedback structure is constructed. By imitating the feedback behavior in the human brain, the operating mechanism is presented. By comparing the recognition result character with the input character, four kinds of general characters recognition errors are defined. According to the analysis of the general errors, the evaluation mechanism and regulation mechanism of closed-loop based on feedback is made. Based on the trend of errors, the recognition process is adjusted to make the whole operating mechanism more reasonable. The results of simulation show that this system is efficient.

A Dynamically Reconfigurable FPGA-Based Pattern Matching Hardware for Subclasses of Regular Expressions

SUMMARY In this paper, we propose a novel architecture for largescale regular expression matching, called dynamically reconfigurable bitparallel NFA architecture (Dynamic BP-NFA), which allows dynamic loading of regular expressions on-the-fly as well as efficient pattern matching for fast data streams. This is the first dynamically reconfigurable hardware with guaranteed performance for the class of extended patterns, which is a subclass of regular expressions consisting of union of characters and its repeat. This class allows operators such as character classes, gaps, optional characters, and bounded and unbounded repeats of character classes. The key to our architecture is the use of bit-parallel pattern matching approach, in which the information of an input non-deterministic finite automaton (NFA) is first compactly encoded in bit-masks stored in a collection of registers and block RAMs. Then, the NFA is efficiently simulated by a fixed circuitry using bitwise Boolean and arithmetic operations consuming one input character per clock regardless of the actual contents of an input text. Experimental results showed that our hardwares for both string and extended patterns were comparable to previous dynamically reconfigurable hardwares in their performances.