33 results about "Stroke order" patented technology

The invention relates to a stylized font generation method based on a diffusion model, and aims to solve the problems of font stroke structure disorder and style deviation in the prior art and realize accurate and stylized font generation. The method is divided into two stages: a first stage, performing pre-training of a font style classification task, including constructing a font style label, designing a style feature extraction network, and training by using a data set to obtain a pre-trained style encoder so as to extract font style features; and in the second stage, training of the generation model is started, style encoder parameters are frozen in the process, content features, stroke order structure features and style features serve as conditions, model parameters are optimized through mean square loss and contrast loss, and therefore a high-quality generation image is trained and generated. According to the method, the structural accuracy and style expressive force of font generation are remarkably improved, and efficient technical support and application value are provided for font library construction and design.

The invention discloses a method and a system for correcting wrong words in minority language text image recognition, and the method comprises the steps: obtaining a minority language text image, removing the high-frequency noise of the minority language text image through employing a preset adaptive threshold denoising algorithm based on wavelet transformation, and generating a denoised image. The morphological contour features of at least one character are extracted from the denoised image, the morphological contour features comprise stroke thickness, stroke number and stroke order direction, the features are input into a support vector machine classification model, the language category of the denoised image is determined, and a language classification result is generated. And loading a corresponding character segmentation model and a corpus according to a language classification result, carrying out character-by-character segmentation and pattern matching identification on the denoised image, and generating a preliminary identification result and an identification confidence score of each character. According to the method for correcting the wrong words in the text image recognition of the minority language, the accuracy of character recognition of the minority language and the capability of correcting the wrong words are improved, and recognition challenges in complex scenes can be coped with.

The invention relates to the technical field of character processing, in particular to a Chinese character stroke extraction method based on domain self-adaption, which comprises the following steps of: expanding stroke data corresponding to Chinese characters in a standard character library; encoding the stroke order priori knowledge of each Chinese character in the standard character library to obtain a stroke splicing graph with stroke order information, and taking the stroke splicing graph as a stroke label of the standard character library; taking each Chinese character in the standard character library and the corresponding stroke spliced graph as source domain data, taking the unlabeled handwritten Chinese character as target domain data, and performing field adaptive training on a pre-constructed stroke extraction model by utilizing the source domain data and the target domain data; and inputting the handwritten Chinese character and the coded stroke splicing graph into the trained stroke extraction model, and outputting a multi-channel stroke image corresponding to the number of strokes of the handwritten Chinese character. According to the method, the accuracy and robustness of handwritten Chinese character stroke extraction can be improved.

The invention discloses a three-view intercommunication calligraphy practicing method based on golden dot calligraphy practicing lattices. The golden dot calligraphy practicing lattices are combined with the three-view intercommunication method to achieve the calligraphy practicing effect. The golden point calligraphy practicing grid comprises square frame lines, the middle points of the adjacent sides of the square frame lines are connected through dotted lines to form an inscribed square auxiliary frame line, the middle points of the opposite sides of the square frame lines are connected through dotted lines to form a cross-shaped auxiliary line, and a golden section point I is arranged on the upper portion of the vertical line of the cross-shaped auxiliary line. A golden section point II is arranged at the left part of the horizontal line of the cross-shaped auxiliary line; the three-view intercommunication method is divided into three steps and nine steps. According to the three-view intercommunication calligraphy practicing method based on the golden point calligraphy practicing grid, through scientific visual space modeling, sensory collaborative training and a multi-dimensional evaluation system, the structural accuracy, stroke order logicality and line quality of Chinese character writing are improved; the training efficiency is high; the method is suitable for calligraphy education, children enlightenment education and adult hard-tipped pen / soft-tipped pen calligraphy improvement scenes.

A computer-readable recording medium stores therein a learning support program that causes a computer to execute a process including: receiving input of a character that a user is requested to write, the character being handwritten and input by the user; recognizing the input character or strokes of the input character; judging the recognized character or a stroke order of the strokes of the recognized character; and outputting information concerning a result of the judging.

The invention discloses a system and a device for identifying names of other people selected from elected handwritten Chinese characters, and belongs to the field of identification, and the method comprises the following steps: analyzing a time dependency relationship of a stroke sequence by adopting a recurrent neural network, and combining a predefined Chinese character stroke sequence rule to obtain a preliminary Chinese character candidate set; for the weighted Chinese character candidate sequence, if the highest confidence is lower than a preset threshold value, analyzing the semantic matching reasonability of the weighted Chinese character candidate sequence through a pre-trained language model to obtain an optimized Chinese character candidate sequence; according to the optimized Chinese character candidate sequence, a sequence matching algorithm is adopted to calculate the similarity between the optimized Chinese character candidate sequence and names in a name database, and a preliminary name candidate set is obtained; and for the final name candidate set, adopting a multi-modal verification algorithm, fusing the repaired stroke feature vector and semantic matching reasonability, determining the confidence coefficient of the final name candidate set, and obtaining a confirmed name record.

The invention discloses a structural analysis literacy method, which comprises the following steps of: (a) literacy analysis: activating understanding through object association, context sentence making and culture traceability, and demonstrating in combination with an ancient character form and a wooden teaching aid; (b) disassembling strokes: decomposing a stroke sequence by using a magnetic gradient teaching aid, and triggering an acousto-optic alarm when the stroke order is wrong; (c) combined writing: training in three stages of hard pen and soft pen, and performing standard detection by means of stroke sticky notes and standard character grids; and (d) multi-modal strengthening: consolidating memory through activities such as puzzle and relay match, and carrying out personalized review based on the wrong character file. According to the technology, a gradient magnetic teaching aid, a three-dimensional evaluation matrix and multi-modal training are integrated, so that the time consumed for mastering a single character is shortened from 12.5 minutes to 6.8 minutes, the stroke error rate is reduced from 18.3% to 4.1%, the repeated practice frequency is reduced by 59% compared with that of a traditional tracing method (5.3 times reach the standard), the teaching efficiency is remarkably improved, and the GB / T 30240-2013 standard requirement is met.

The present application belongs to the field of computer vision and vector image generation, and particularly relates to a vectorized Chinese character pattern generation method based on a large model. The method comprises the following steps: step one, data preprocessing; step two, stroke feature extraction; step three, feature vector discretization coding; step four, stroke sequence combination; step five, feature coding prediction; step six, coding feature restoration; and step seven, stroke restoration. The present method combines Chinese character stroke order information, and on the basis of being able to complete Chinese character stroke generation to form a complete vector image, can also generate continuous semantic Chinese character vector text content in the form of continuously predicting the next character stroke, can complete missing character shapes in the case of incomplete font library, realize creative vectorized Chinese character text generation, make contributions to the fields of artistic character shape generation, Chinese character font auxiliary design, advertisement text design, etc., reduce labor costs, and realize efficient generation.

The application discloses a Chinese character sorting, searching and dictionary constructing method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the following steps: obtaining a first character of a Chinese character as a first character according to a stroke order, and encoding the first character to obtain a first character code, wherein the first character comprises a first character component which can be independently formed into a character and cannot be split, and a first main stroke component of the Chinese character which does not comprise any character component; obtaining the number of strokes included in each Chinese character, and obtaining a stroke string corresponding to each Chinese character according to a preset stroke character coding rule; sorting the Chinese characters by taking the first character code as a first sorting field, the number of strokes as a second sorting field, and the stroke string as a third sorting field, wherein the first sorting field has the highest priority, and the third sorting field has the lowest priority. The embodiment can solve the problems of sorting and searching difficulties caused by the definition and position of the component.

Systems or methods for presenting or generating an animated realization of one or more letter glyphs based on progressively exposing a continuous mapping of brush head shapes following a letterform stroke order. A machine-learning model is used to generate animation metadata from the one or more letter glyphs. The animation metadata including a centerline, the continuous mapping of brush head shapes that is used to sweep along the centerline, and the letterform stroke order to generate the animated realization of the one or more letter glyphs.

The present application provides a method, apparatus and device for digital processing of text, which relates to the technical field of information processing, and can reduce the distortion rate of the digital standard point set in text restoration, and improve the restoration degree and aesthetic effect of text. The method includes: obtaining first skeleton information of the target text, where the first skeleton information is a plurality of pixel points extracted from the target image, and the target image is an image corresponding to the target text; correcting the pixel points with errors in the first skeleton information to obtain second skeleton information; performing stroke division and stroke order restoration on the second skeleton information to obtain first stroke order information; and performing widening processing on the pixel points in the first stroke order information according to the line width information to obtain the standard point set of the target text.

The present invention discloses a novel pure pen-shaped Chinese character input method and its special keyboard. The input method includes the following steps: Step 1, encode the root characters on the key positions of the information processing device. Among them, the types of root characters include radical root characters, whole-character root characters, stroke root characters, and combined root characters. The encoding of radical root characters is mainly the initial letter of the pinyin of the common name of the radical, and the encoding of whole-character root characters is mainly the initial letter of the pinyin of the character; Step 2, use various said root characters to split the Chinese character to be input according to the stroke order of the Chinese character, and select at most four root characters from the split root characters; Step 3, type the keys corresponding to the at most four root characters, thereby inputting the Chinese character. The special keyboard includes 13 or 14 letter keys, and 26 letters are distributed on the 13 or 14 letter keys. This input method has the characteristics of simple rules, direct image, easy to remember, and can effectively improve the ability of information processing devices such as computers and mobile phones to process Chinese characters, etc.

The application discloses a method and a terminal for automatically generating handwriting Chinese character animation, obtains a target Chinese character for which handwriting Chinese character animation is to be generated, analyzes each stroke of the target Chinese character and corresponding stroke type and center point data of the stroke; determines corresponding vertex data according to the stroke type of the stroke; calculates connection data required for drawing the stroke according to the center point data and the vertex data corresponding to the stroke; generates handwriting stroke animation of the corresponding stroke according to the connection data, integrates handwriting stroke animation corresponding to each stroke according to the stroke order of the target Chinese character, and obtains handwriting Chinese character animation of the target Chinese character. The application can automatically generate handwriting Chinese character animation according to the requirements of a user, simulate a real writing process during generation, improve the interactive experience of the user, and save manpower and resources.

This invention discloses an automatic pattern-making method for Chinese character embroidery based on the fusion of semantic-driven and process-constrained approaches, solving the problems of unreasonable stitch distribution, severe path redundancy, and difficulty in balancing semantic structure and physical manufacturing constraints in existing automatic embroidery pattern-making technologies. The method includes: acquiring unstructured Chinese character images; performing target detection and semantic recognition on the Chinese character regions to generate structured semantic labels; introducing standard stroke order templates based on the semantic labels and decoupling the topological structure of the Chinese characters through deformation registration to obtain single-stroke processing units arranged in the writing order; performing geometric modeling on the single strokes, extracting the central axis, and reconstructing continuous boundaries; performing topological partitioning of complex stroke regions based on pivot decomposition to generate sub-regions suitable for embroidery processing; generating adaptive stitch paths within the sub-regions and optimizing the paths; and finally quantizing and encoding the stitch trajectories to generate processing instruction files that can be executed by embroidery equipment.

The present application relates to a method and system for intelligent assessment of Chinese character writing ability, comprising the following steps: when a recognition start command is received, writing data sent by a writing recognition terminal is received in real time; the writing data is sent to a writing analysis model, the writing analysis model compares the stroke order data of the Chinese character text with pre-stored standard stroke order information, and outputs a stroke order comparison result; the stroke order comparison result includes correct stroke order information and incorrect stroke order information; the Chinese character text is compared with a preset standard text for contour similarity, and a comparison score is output; if incorrect stroke order information exists, the Chinese character text associated with the incorrect stroke order information is screened out, and the standard stroke order information corresponding to the Chinese character text is associated with the Chinese character text to generate a stroke order correction data packet; the comparison score and the stroke order correction data packet of the Chinese character text are sent to the writing recognition terminal. The present application has the effect of facilitating the judgment of the correctness of each student's writing stroke order.

The invention relates to a handwriting pen order detection method and device, and the method comprises the steps: carrying out the calibration of a binocular camera based on a calibration plate disposed at the end part of a handwriting pen, so as to obtain a continuous image of the writing process of the handwriting pen; calculating the track of the handwriting pen in real time based on the pixel coordinate change of the geometric center of the calibration plate in the binocular camera; segmenting the track based on the normal motion characteristics of the calibration plate; based on the trained neural network, performing incremental accumulation recognition on each segmented track until a Chinese character is recognized; based on a preset Chinese character stroke order database, calculating the similarity between each identified Chinese character stroke graph and a standard stroke to obtain a stroke order code; and judging whether the writing sequence is correct or not according to the stroke order codes and the stroke order codes corresponding to the recognized Chinese characters. Through binocular camera detection and the neural network, stroke order detection of the handwriting pen is realized, the method can be applied to a daily writing homework scene, and more efficient writing standard training is realized.

The invention discloses an intelligent anti-shake pen based on vector force guidance and a writing method, and the intelligent anti-shake pen comprises a pen body housing, an internal core support, a slidable writing refill assembly, a tremor suppression module, a vector force stroke auxiliary driving module, a bidirectional interaction module and a main controller. The tremor suppression module monitors the tremor transmitted to the pen body due to the tremor of the hand of a user in real time through a built-in inertial measurement unit and drives the reverse micro-vibration assembly to generate dynamically matched compensation torque, and meanwhile, the vector force stroke auxiliary driving module outputs the compensation torque to the pen body according to a stroke instruction issued by an external intelligent terminal. Physical guiding force which is controllable in direction in the two-dimensional writing plane is applied to the writing refill assembly. And the main controller cooperatively regulates and controls anti-shake and guide, and performs closed-loop communication with an external terminal through the bidirectional interaction module. According to the invention, active anti-shake and active stroke guide functions are combined, so that a hand tremor patient can be effectively helped to write stably, a writing learner is assisted to master a correct stroke order, and the writing stability and accuracy are remarkably improved.

In order to solve the problem that an existing input method cannot have both easy learning and high efficiency, the invention creates a Chinese character self-assembling Chinese input method and a character root keyboard thereof which can freely select roots, autonomously split characters and self-automatically encode, establish Chinese character shape codes and efficiently typewrite on the basis of general common knowledge of the public. The method is suitable for the public cognitive common knowledge, more than 770,000 character splitting and coding schemes and related 3196 kinds of etymon and keyboard key mapping relations are created for more than 20,000 GBK Chinese characters, a user does not need to learn any etymon table, etymon songs, etymon key mapping maps, key names, fonts, character splitting and coding specifications or special rules, and the method is simple and convenient. A user can easily print wanted characters and words only by thought of the existing cognition of Chinese character structures, stroke orders, etymons and pronunciation of the Chinese characters and thought of splitting and coding; through a quantitative analysis algorithm, a character splitting and coding scheme of Chinese characters is preferentially graded and sequenced, and a user can input all GBK Chinese characters and 600,000 phrases in a touch typing manner by applying a simple coding skill without turning a screen.

The invention mainly relates to the technical field of software development, and provides a method and system for drawing Chinese characters based on a Qt platform in order to improve the visualization degree of a Chinese character writing rule in a Chinese character teaching scene, and the method and system are characterized in that the Qt platform loads character contour data and stroke node data of corresponding Chinese characters according to target Chinese characters input by a user; fitting Chinese character contours based on the character contour data; according to the method, the filling spacing is calculated based on the stroke node data, and color filling is performed on the interiors of the Chinese character contours according to the calculated filling spacing, so that cross-platform visual consistency during Chinese character writing is ensured, dependence on a system font engine is eliminated, and advanced teaching functions such as stroke order animation, interaction highlight and radical decomposition are endowed to a Chinese character display unit.

The invention provides an online handwriting stroke order feature construction method and system and a handwriting identification method, and the method comprises the steps: analyzing handwriting collection data (x, y, p, t, s), and segmenting the handwriting into a plurality of strokes by using a point with a speed v close to 0 as a segmentation point; constructing a stroke order feature S including a stroke order feature, a stroke angle feature, a stroke type feature, a stroke length feature, a stroke position feature and a stroke conversion feature; and processing the stroke order feature S through frequency mapping transformation, extracting a high-frequency spectrum component in the stroke order feature S by using a high-frequency oscillation primary function in the frequency mapping transformation, and converting the stroke order feature S into a representation form suitable for a deep learning model. The method not only can reflect the sequence of strokes, but also can reflect the space structure difference caused by the sequence of strokes, remarkably improves the problem that a deep learning model is insensitive to different stroke orders, and improves the performance and generalization ability of the model.

The invention discloses a concave finger mark stroke literacy and calligraphy practicing card and a method thereof, belongs to the technical field of Chinese character literacy and calligraphy practicing auxiliary tools, and solves the technical problems that an existing calligraphy practicing tool lacks finger transition practicing, literacy and calligraphy practicing are separated, and adaptability is poor. According to the calligraphy practicing card, a hard material serves as a base material, the base material is provided with matts, stroke pits simulating finger writing traces are formed in the matts, auxiliary character learning information such as syllables and stroke orders is printed, and the card body size and content can be customized according to different crowds; according to the matched calligraphy practicing method, muscle memory formed by finger depicting and sinking serves as the core, and seamless connection from finger depicting to pen writing is achieved through the steps of card selection, depicting, character learning, pen writing transition and the like. The device is low in manufacturing cost, can effectively standardize the hand writing track, reduces the calligraphy practicing difficulty, achieves the synchronization of character learning and calligraphy practicing, is suitable for children, students, foreigners and other groups, is easy to popularize in batches, and greatly improves the Chinese character learning and writing practicing efficiency.

The invention relates to the technical field of Chinese character teaching, in particular to a Chinese character writing stroke order real-time error correction system and method based on deep learning, and the system comprises a preprocessing module, a multi-scale space-time convolution network module, a stroke order error diagnosis module and a correction feedback module. The multi-scale space-time convolutional network module extracts space-time features through a double-flow heterogeneous architecture, constructs a feature pyramid, and enhances key features by applying a dynamic attention mechanism, and the stroke order error diagnosis module analyzes stroke order features based on a rule base, detects multiple error types through multi-task learning, and determines the stroke order error. According to the system, 42 stroke order errors can be accurately recognized, targeted feedback is provided, and Chinese character writing teaching is effectively assisted.

The invention discloses a handwritten Chinese character stroke splitting method and device, equipment, a medium and a product, and relates to the technical field of image processing, and the method comprises the steps: carrying out the recognition and extraction of a handwritten Chinese character in an obtained target image, and obtaining a target Chinese character and a recognition result of the target Chinese character; according to the recognition result of the target Chinese character, obtaining and temporarily storing a target stroke matrix of the target Chinese character from a Chinese character stroke order library; according to the stroke sequence of each target stroke matrix of the target Chinese character, sequentially executing stroke extraction operation on each target stroke matrix of the target Chinese character, extracting target strokes, and obtaining a stroke splitting result of the target Chinese character; after executing the stroke extraction operation each time, if the stroke sequence of the current target stroke matrix is greater than 1 and the orientation of the target historical stroke in the target Chinese character is inconsistent with the orientation of the target historical stroke in the current stroke splitting result, performing orientation exchange on the current target stroke and the target historical stroke; the stroke splitting effect is improved.

To provide a functionality of handwriting input in a user device.SOLUTION: A handwriting recognition module is trained to have a repertory including a plurality of non-duplicate scripts, and can use a single handwriting recognition model to recognize tens of thousands of characters. A handwriting input module provides real-time stroke order and stroke direction independent handwriting recognition relating to handwriting input for plural characters. Specifically, the real-time stroke order and stroke direction independent handwriting recognition is provided relating to Chinese handwriting recognition at a level of plural characters or sentences. A user interface for providing a functionality of the handwriting input is also disclosed.SELECTED DRAWING: Figure 8A

The present disclosure relates to a stroke detection method and device, readable medium and electronic equipment. The method comprises: obtaining inputted to-be-recognized strokes in a user writing a target character, the target character being composed of at least two standard strokes; determining a similarity between the to-be-recognized strokes and a target standard stroke as a first similarity probability, the target standard stroke being at least one of the standard strokes; determining a probability that the to-be-recognized strokes are the target standard strokes as a second similarity probability according to input content before the to-be-recognized strokes; and determining whether a stroke order of writing the to-be-recognized strokes is correct according to the first similarity probability and the second similarity probability. Thus, the accuracy of stroke detection can be effectively improved, the efficiency of stroke detection can be improved, and real-time detection can be realized during user writing.

The present invention relates to computer technology, and specifically to a method for identifying illegal advertisements that combines text visual features and text content features. The method utilizes three methods to convert a color image into three features: the text visual image itself, the text visual image stroke features, and the image text recognition into text features. VGG16 is used to extract image features. RAN is used to extract the stroke order features of Chinese characters. A CNN-encoder / decoder architecture is used to extract the stroke features of Chinese characters and then embeds and transforms them into image features. DenseNet is used for feature learning and classification. Optical Character Recognition (OCR) is used to convert the text portion of an image into text information. Rule matching and review are performed on the text information. A RoFormer neural network model is used to perform feature learning and splitting on sentence-level text information. The method then identifies illegal advertisements. This method can provide strong support for the advertising review and supervision systems of major website platforms.

The application discloses a method and a terminal for generating a handwriting Chinese character animation quickly, acquires a target Chinese character for which a handwriting Chinese character animation is to be generated, selects matching data corresponding to the target Chinese character in a Chinese character database, generates a connecting point set of each stroke including a starting vertex and an ending vertex according to the matching data, applies a connecting line matching rule to the connecting points in the connecting point set for connecting line matching to generate connecting line data of the corresponding stroke, and generates a handwriting Chinese character animation of the target Chinese character according to stroke orders of all strokes of the target Chinese character and the connecting line data. The application can acquire connecting line related information corresponding to the target Chinese character more accurately and quickly based on data matching, generates connecting line data of each stroke in real time, simulates real handwriting, realizes strokes of Chinese characters appearing one by one, helps learners to understand and practice Chinese character writing better, has a more real and vivid experience, saves storage space, and is not limited by storage or performance.

The invention discloses a handwritten Chinese character image evaluation system based on deep learning, and belongs to the field of Chinese character image evaluation, and the system comprises a preparation processing module which carries out Chinese character stroke sequence extraction operation on a Chinese character image to be written, obtains a Chinese character skeleton image, carries out segmentation operation on Chinese character strokes, and obtains a Chinese character skeleton image; and an explosion image containing a plurality of pencils is obtained. According to the method, the step correction module guides the writing position of the user, so that the Chinese characters written by the user are regular, convenience is brought to the user, meanwhile, the recognition rate is increased, the step correction module does not limit the writing range of the user, the problem that part of handwriting is cut off is effectively avoided, and the writing smoothness is improved; through the preparation processing module, the Chinese character strokes are subjected to segmentation operation in advance, the explosion image containing a plurality of strokes is obtained, the step of Chinese character segmentation is omitted, and identification errors caused by wrong segmentation are avoided.

The invention provides a handwritten Chinese character stroke order recognition method based on a semantic segmentation model, and the method comprises the steps: converting handwritten sequence data into a handwritten Chinese character path grayscale image Ih, and converting printed sequence data into a printed path grayscale image Ip; predicting a spliced image of a handwritten Chinese character path gray level image Ih and a printed form path gray level image Ip by using a semantic segmentation model, wherein the model takes a dual-channel gray level image as input; generating a mask pattern of each standard stroke according to the printing body sequence data; calculating an average gray value of each standard stroke mask image on the spliced image predicted by the model; and sorting the average gray values of all the standard strokes so as to determine a stroke writing sequence and realize stroke order identification. By applying the technical scheme of the invention, the problem of difficulty in accurately and efficiently identifying the stroke order of handwritten Chinese characters with various writing styles and non-standard stroke orders in the prior art can be effectively solved.

The application discloses a writing evaluation method, device and equipment, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring stroke data of a to-be-tested font in real time; for a pre-set stroke order evaluation dimension, determining the difference between each to-be-tested stroke of the to-be-tested font and each standard stroke of a standard font according to the stroke data of the to-be-tested font and the stroke data of the standard font; determining the to-be-tested stroke that is matched with each standard stroke of the standard font one by one among each to-be-tested stroke of the to-be-tested font according to the difference, and forming an index pair by combining the first stroke order index of each standard stroke with the second stroke order index of the to-be-tested stroke matched with each standard stroke; and generating a writing evaluation result of the to-be-tested font in the stroke order evaluation dimension according to the index pair. The application can effectively evaluate the case that the writing is correct but the stroke order is wrong in the writing process of children.