55 results about "Cursive" patented technology

The invention discloses an automatic generation method for a handwritten Chinese font library based on a deep neutral network. The method is characterized in that font characteristics are reconstructed through a small number of handwritten Chinese characters; a font style migration network is established; and font style characteristics of font forms not handwritten by a user are estimated throughthe font style migration network, font form contents of a reference font are combined with a handwriting style of the user to achieve migration to a target handwriting style, and target font form pictures are generated, so that the complete handwritten Chinese font library is obtained. The method provided by the invention has the advantages that the method is an end-to-end generation method, and high-quality Chinese font forms are generated without extraction of strokes or components of Chinese characters or manual intervention, so that the handwritten font library creation efficiency is greatly improved, the generation of a personalized font library becomes simple and convenient, the requirement of ordinary people for the personalized handwritten font library can be met, and the development progress of the personalized font library can be accelerated.

A cursive character handwriting recognition system includes image processing means for processing an image of a handwritten word of one or more characters and classification means for determining an optimal string of one or more characters as composing the imaged word. The processing means segments the characters such that each character is made up of one or more segments and determines a sequence of the segments using an over-segmentation-relabeling algorithm. The system also includes feature extraction means for deriving a feature vector to represent feature information of one segment or a combination of several consecutive segments. The over-segmentation-relabeling algorithm places certain segments considered as diacritics or small segments so as to immediately precede or follow a segment of the associated main character body. Additionally, the system also includes classification means that processes each string of segments and outputs a number of optimal strings which could be matched against a given lexicon.

The invention provides a hand-written Chinese character beautification method, which is a handwriting processing method for hand-written Chinese characters on the basis of retaining the writing style of the writer. The invention aims at the input trail captured by general input devices (such as mouse, and touch-screens without pressure-level induction) and carries out speed-based calligraphy stroke simulation, stroke matching and template matching to the input trail so as to achieve the purpose of beautifying the writing trail. The invention also provides a personalized hand-written trail processing way which enables people who lack of specialized calligraphy practice to create beautiful handwritten calligraphic Chinese characters with personal style.

The invention discloses a Chinese word stock automatic generation method based on writing style modeling, and the method comprises the steps: carrying out the stroke track automatic extraction and wrong result screening operation of inputted handwritten Chinese character image, and obtaining training data; carrying out the learning and modeling of handwritten Chinese character stroke shape and style and stroke inter-frame structure style through a neural network, and obtaining handwriting stroke connection characteristics and stroke end part contour handwriting features through statistical analysis, and then generating a high-quality Chinese character form consistent with the handwriting style of a user. The method just needs a few of commonly-used Chinese characters as the input, does not need any manual intervention, can automatically generate a vector Chinese word stock comprising a large number of Chinese character patterns, can quickly and automatically generates a handwriting Chinese word stock with other handwriting style for a user, can remarkably improves the manufacturing efficiency of the Chinese word stock, and greatly reduces the production cost.

The invention discloses a method for generating and beautifying handwritten characters with a computer. The method is realized by learning a certain amount of hand written character samples. The method comprises the following steps: decomposing and parameterizing the strokes of five to twenty different handwritten character samples for the same character, and converting each character into the form of sample vector by the image processing method; linear interpolation is conducted among characters to obtain new handwritten characters; evaluating each new character by an automatic computer evaluation method of character writing form aesthetic degree; and adjusting interpolation coefficients repeatedly through a nonlinear programming optimization method so as to achieve the effect of beautifying the character. The system of the invention can automatically generate handwritten characters with better aesthetic degree, and improve the aesthetic degree of handwritten characters of a user on the basis of reserving the basic characteristic of the handwriting of the user.

The invention discloses an off-line handwritten form Chinese character recognition method and belongs to the technical field of Chinese character recognition and extraction. The method comprises the steps of data collection, preprocessing, feature extraction and recognition extraction. A classification and recognition standard in the Chinese character recognition method comprises that a sample to be tested is classified according to a minimum distance standard, the feature recognition is performed based on a 4*4 elastic mesh, in rough classification recognition, the number of feature points such as end points, inflection points and cross points in each of meshes are conducted statistics on, and weight sum is respectively performed, in a second classification recognition, the number of transverse, vertical, left-falling and right-falling stroke vectors in each of the meshes is recorded, and handwritten form Chinese characters are recognized and extracted. According to the method, both off-line handwritten form Chinese characters and print form Chinese characters can be recognized, the feature extraction method of the Chinese character recognition method is simple, and the recognition speed is fast.

The invention relates to a method that can directly write in handwritten information, belonging to information input technology, which is characterized in that: no information identification and selection process is needed; the information is segmented or combined into a plurality of independent information units, and then the original handwritten handwriting is directly displayed on a display component; the operator writes in the handwriting information by an input member; the system captures the handwriting information write in by the operator and segments or combines the handwriting information into independent information units, and then the original handwriting is displayed on the display component. The invention has the advantages that the handwriting input by the method can further facilitate edit operation; the method is more suitable for the input of Chinese characters, special characters and formulas, and the information input efficiency is greatly improved; meanwhile the original handwriting of the writer is retained, and common informal documents can have more individualized characteristics.

The invention discloses a printed and handwritten mixed text line extraction system, which comprises text block area preprocessing and text line extraction. Through the text block area preprocessing process firstly, a document image containing a printed or handwritten or mixed text is subjected to skew correction, and a text block area is acquired through layout analysis; the text block area is then subjected to connected domain clustering, the belonging text line label is distributed, touching character pixels are subjected to clustering segmentation and the belonging text line label is redistributed; and finally, through the text line label, extraction of a document image text line is realized. The defects of missing extracted stroke information or redundant stroke information introduction caused by inclined touching of the text line in a handwritten text or a printed and handwritten mixed text in the prior art can be overcome, the pixel controllability during the text line extraction process is improved, and the purpose of quickly and efficiently extracting the text line is thus achieved.

The invention provides a rejection method for identifying handwritten character based on multiple classifiers, belonging to the field of handwritten character identification. The rejection method is led into a handwritten character identifying system, so as to effective improve the reliability of the identifying system and improve the identifying rate of the identifying system. The invention provides a rejection method based on a single classifier and the rejection method based on the multiple classifiers. Aiming at the limitation of a single characteristic rejection method, the idea of a multiple classifiers system is adopted to design the single classifier for the each characteristic, and integrate the rejection result of the each classifier, so as to play advantages of the each characteristic and improve the reliability of the rejection. Based on the rejection method of the single classifier, the invention provides the rejection method of the multiple classifiers. The method can better solve contradiction between the identifying rate and the rejecting rate of the handwritten character identifying system, and improve the reliability of the identifying system.

Input handwritten characters are classified as print or cursive based upon numerical feature values calculated from the shape of an input character. The feature values are applied to inputs of an artificial neural network which outputs a probability of the input character being print or cursive. If a character is classified as print, it is analyzed by a print character recognizer. If a character is classified as cursive, it is analyzed using a cursive character recognizer. The cursive character recognizer compares the input character to multiple prototype characters using a Dynamic Time Warping (DTW) algorithm.

Input handwritten characters are classified as print or cursive based upon numerical feature values calculated from the shape of an input character. The feature values are applied to inputs of an artificial neural network which outputs a probability of the input character being print or cursive. If a character is classified as print, it is analyzed by a print character recognizer. If a character is classified as cursive, it is analyzed using a cursive character recognizer. The cursive character recognizer compares the input character to multiple prototype characters using a Dynamic Time Warping (DTW) algorithm.

The invention discloses a style character generation method based on a small number of samples. A style reference character data set composed of several style characters (handwriting style or print style) and a character of a standard font are used as a character content prototype data source, and an image translation model based on a deep generation antagonism network is used to train a charactergeneration model of character style transfer. The model can generate arbitrary characters with the same writing / printing style by using a given few (or even one) characters with a certain style (writing / printing) as the style reference template. The content of the generated character is determined by the typed content prototype (standard font).

The invention discloses a method for identifying handwritten Uigur characters, which belongs to the field of character mode identification and can effectively identify 128 classes of handwritten Uigur characters in real time. The method comprises the following three parts of: (1) establishment of a Uigur character component dictionary and a handwritten Uigur character component library; (2) a character training process, comprising character one-line information acquisition, character pre-processing, component partition during training, extraction of time division direction characteristics, calculation of training characteristics of components and the like; and (3) a character identification process, comprising character one-line information acquisition, character pre-processing, componentpartition during training, extraction of time division direction characteristics and point component characteristics, characteristic fusion of components, identification result output and the like. The method is the basis for identifying whole Uigur words. By using the method and a handwritten Uigur word splitting method, a user can freely, naturally and smoothly write the Uigur words at random angle in a mobile platform handwriting frame such as an intelligent mobile phone and the like, and the method can realize robust identification.

The invention discloses a method and a system for generating a wrong answer list. The method comprises the following steps of: acquiring an electronic original test question picture, wherein the electronic original test question picture comprises printed characters and handwritten characters, and the colors of the printed characters are different from those of the handwritten characters and are consistent with those of characters in actual test paper; recognizing a color value of each pixel point in the electronic original test question picture; modifying colors of the pixel points of which the color values are in a preset range into background colors, and saving as a test question picture which does not contain the handwritten characters; and exporting the test question picture which does not contain the handwritten characters and the corresponding original test question picture to two word documents respectively to generate the wrong answer list, wherein the test question pictures in the two word documents are required to correspond one by one. Through the method and the system, the efficiency of generating the wrong answer list can be improved.

The invention provides a method for imitating handwriting of Chinese characters by using a computer. A computer image processing and artificial intelligent method can be adopted for learning and imitating personal handwriting with a computer. In the method, the available personal Chinese character samples in handwritten form are resolved in strokes and classified; each stroke, each radical, a character and the inner space structure of the character are parameterized in a unified way, and a personal handwritten character font database is established in combination with a plurality of ordinary types; when the personal handwriting is imitated, the Chinese character to be imitated is resolved into a radical or strokes, and each radical or stroke is found from the personal handwritten character font database one by one; the radical or the stroke is reconstituted according to the inner space structure features of the personal character as the output of the imitated handwriting. In the method, the stroke radical constitution rules of Chinese characters are analyzed directly; the characteristics of personal handwriting are seized more essentially, and personal handwriting is imitated better.

The invention discloses a method and a system for automatically creating error homework textbooks. The method includes acquiring electronic-edition test question pictures which consist of texts including print-form texts and handwritten texts; extracting text characteristics from the test question pictures; comparing the extracted text characteristics with a preset text characteristic library, and identifying the print-form texts and the handwritten texts; and erasing the handwritten texts from the test question pictures, generating test question pictures without the handwritten texts, and accordingly automatically creating the error homework textbooks according to the test question pictures containing the handwritten texts and the test question pictures without the handwritten texts. By the aid of the method and the system, the test question pictures without handwritten texts can be created according to the electronic-edition test question pictures, and furthermore, the error homework textbooks can be automatically created.

The invention discloses a character identification method and a character identification system. The character identification method comprises steps of inputting a to-be-identified area image and performing binary processing; extracting stroke characteristics of the to-be-identified area image; extracting relation characteristics between adjacent strokes of all strokes and the corresponding positions; matching radicals; matching characters; and outputting matching results of the to-be-identified area image. The character identification system comprises an image acquiring module, an image processing and character identifying module and a character outputting module. According to the invention, printed and quite neatly handwritten characters can be precisely identified; there are no other restrictions except that the handwritten characters could not be too illegible to read the structure characteristics; and a large number of printed and handwritten manuscripts, tables and bills can be processed at a high speed.

The invention discloses an offline handwritten mathematical formula recognition and reconstruction method, which is characterized in that a character segmentation problem is converted into a target detection problem in a formula segmentation stage, so that the segmentation problem of broken characters and adhesion characters is effectively solved, and the accuracy of a character segmentation result is greatly improved; at the character recognition phase, two convolutional neural network models based on batch normalization and global average pooling methods and only using convolution kernels with the size of 3 * 3 are designed for the problem of offline handwritten mathematical character recognition, the voting results of the two models are used as recognition results in the recognition stage, so that the recognition of mathematical characters can be accurately completed, and the character recognition accuracy is improved; at the formula analysis and reconstruction stage, the geometricconstraint is carried out according to the position relation among the characters, the semantic constraint is carried out by using a two-dimensional random context-independent grammar, and finally, the characters are combined from bottom to top, so that the formula reconstruction is completed.

The invention discloses a handwriting separation and positioning method based on a printing template. The handwriting separation and positioning method specifically comprises the following steps: (1)image correction: carrying out binarization processing, contour extraction and Hough transform on a blank template and an image containing a handwriting text to detect and correct the inclination of the image; (2) handwriting separation: separating the handwriting and the printing form in the picture containing the handwriting text to obtain a pure handwriting picture; (3) text positioning: positioning the handwritten text in the obtained pure handwritten picture; (4) straight line detection: detecting whether a straight line exists in the handwritten text positioned in the step (3), and if so, detecting the specific position of the straight line and outputting the coordinate value of the straight line; and (5) target extraction: carrying out interference character screening on the pure handwritten text, and removing the interference handwritten characters to obtain the required handwritten characters. According to the method, the handwritten part containing the printing body and the handwritten picture at the same time is accurately positioned, and the accuracy of the text recognition system is improved.

The invention discloses a computer-assisted instruction method aiming at the writing of Chinese characters. By adopting a series of computer image processing and artificial intelligence, the method can realize automatic or semiautomatic computer-assisted instruction on the writing of Chinese characters. The method comprises the following steps: artificial scoring is performed to the features and the appearance ratings of a great quantity of hand-written character samples in advance; then the hand-written input of a writing user is received, and an automatic computer evaluation method for the appearance ratings of the writing shape of the characters is used for scoring the hand-written characters of the user; the hand-written characters of the user are parameterized and transformed to be expressed by the linear sum of a plurality of standard characters; and finally, a nonlinear programming optimization method is adopted, the interpolation coefficients are repeatedly adjusted to raise the score of the appearance ratings of the characters, and the user is guided to improve the appearance ratings of the characters on the basis of retaining the character features of the user; besides, the guidance can also be performed to only a part of a character or the overall structure of the character with the designation of the user.

The present invention relates to the Chinese character recognizing method based on structure model, and belongs to the fields of mode distinction, artificial intelligence and Chinese character information processing. The present invention constitutes two mathematic models, including stroke section center point model and stroke relational matrix model, for describing Chinese character structures and stroke section center point recognizing method and stroke relational matrix recognizing method with stroke section and stroke primitives, and the models. The Chinese characters are classified roughly in stroke section center point recognizing method and delicately in stroke relational matrix recognizing method, and the present invention forms one integral Chinese character recognizing method for both book letter and script Chinese characters with high correctness and stable performance.

Systems and associated methodology are presented for Arabic handwriting synthesis including accessing character shape images of an alphabet, determining a connection point location between two or more character shapes based on a calculated right edge position and a calculated left edge position of the character shape images, extracting character features that describe language attributes and width attributes of characters of the character shape images, the language attributes including character Kashida attributes, and generating images of cursive text based on the character Kashida attribues and the width attribues.

The invention discloses a soft segmentation method of financial OCR system handwritten numerical strings. Automatic recognition processing of financial paper realizes automatic input and rechecking of financial paper, seamlessly integrates the whole process of image processing, layout analysis and intelligent identification, and comprises automatic classification of financial paper images, image preprocessing of the financial paper, identification, supervision and checking of elements in the financial paper, and the like. OCR technology is the core part in a financial paper automatic recognition processing system, requires cutting connected character strings required for automatic processing of financial paper elements into single characters, and performing character recognition. The character recognizer at the present stage has high accuracy rate, so that the overall recognition rate of the OCR system depends on the accurate rate and the acceptability of character string segmentation. The method aims at solving the technical problem of a soft segmentation method for realizing connected numeric strings based on a fuzzy pattern recognition theory, so as to improve the accurate rate of the integral segmentation flow and reduce the false rejection rate of the system, and improve the overall performance of the recognition system.

The invention discloses a handwritten form keyword detection method comprising that: characteristic points of keyword images in a keyword image database are extracted so that a keyword characteristic database is established; the characteristic points of texts in text images to be detected are extracted so that a characteristic point database of the text images to be detected is obtained; sliding windows of the text images to be detected are extracted, characteristic point sets corresponding to the sliding windows are extracted from the characteristic point database, and the characteristic point sets are compared so that initial matching point pair sets are obtained; the initial matching point pair sets are screened so that accurate matching point pairs are obtained; and the sliding windows in the text images to be detected are screened according to the matching point pairs and integrated so that detection results are obtained. In the process of characteristic extraction, SIFT characteristic representation is adopted. The method is applicable to detection of keywords of large amounts of handwritten form documents, such as historical literatures, correspondences, notes, etc. Under the premise of establishing a black list image database, the keywords in a black list can be effectively detected and handwriting of different authors can also be distinguished.

A progressive handwriting tool utilizes one of (a) color contrast within a writing segment, or (b) a grid of sloped lines, to properly guide a student in writing print and / or cursive letters in a proper style and form. In one aspect, a print writing tool includes a writing medium, such as a sheet of paper or an electronic input capturing device, where a plurality of writing segments comprised of first and second rectangular regions are formed thereon. The rectangular regions extend horizontally in a stacked arrangement, and have contrasting colors to clearly differentiate between upper and lower areas where uppercase and lowercase letters are to be written. In another aspect, a cursive writing tool includes a writing medium onto which is formed a set of parallel horizontal lines and a set of diagonal lines of constant slope. The horizontal and diagonal lines intersect to form an array of writing boxes, with each box configured to have one letter of a cursive word written therein. The progressive handwriting tool may be accompanied by writing programs or instructions on how to practice handwriting skills with the tool, such that the programs and tool together form an instructional writing kit.

The invention relates to a handwritten text editing method which comprises the following steps of: receiving preset size data of a single handwritten text; acquiring stroke data of each input period of handwritten input equipment, wherein the stroke data comprises a stroke trace, starting time and ending time; combining the acquired stroke data of a plurality of periods into a handwritten text group; splitting the handwritten text group into single handwritten texts; zooming the split single handwritten texts at equal proportion according to a preset size; and displaying the zoomed single handwritten texts in an appointed area according to an original sequence. Compared with the prior art, the handwritten text editing method has the advantages that: according to the handwritten text editing method required to be protected, after the handwritten text group is split into the single handwritten texts, the sizes of the single handwritten texts can be normally zoomed like editing the word sizes of general printed texts, so that an aim of averaging the sizes of the handwritten texts.

The invention discloses an English handwriting identification method based on improved VGG-16 model which comprises the following steps of collecting English document data sets from different handwriting of different persons; carrying out word segmentation on the obtained handwritten English handwriting document to obtain a data set composed of English words, and constructing a training sample setand a test sample set based on the data set; obtaining an improved VGG-16 model used for constructing a convolutional neural network, wherein the model comprises two traditional convolutional layers,three composite convolutional layers, five pooling layers and three full connection layers; inputting a training sample set, extracting character features, and carrying out classification training; and automatically identifying the English handwriting by using the trained neural network. The identification of English handwriting can be efficiently realized and identification accuracy rate can reach 100%. An algorithm has certain robustness on illumination change, simple geometric deformation and additional noise, and can be used in related fields of handwriting English handwriting identification. The algorithm can also be used for handwriting identification of other character handwritings after being expanded.

The invention provides a handwritten numeral recognition method and a handwritten numeral recognition device. The method comprises the following steps of: receiving a handwritten numeral image input by a user; extracting M covariance characteristics of the handwritten numeral image, wherein the value of M is any one positive integer which is greater than 1; and recognizing handwritten numerals according to the preset M covariance characteristics of each training image, a preset type tag carried by each training image and the M covariance characteristics of the handwritten numeral image. The handwritten numerals are recognized by extracting the M covariance characteristics of the handwritten numeral image, wherein the value of the M is any one positive integer which is greater than 1, so that the problem of inaccurate recognition of the handwritten numerals caused by adoption of a single covariance characteristic in the recognition process of the handwritten numerals in the prior art is solved.