132 results about "Single character" patented technology

The invention discloses a character loss compensation processing device and method for video subtitle extraction, which effectively solve the problem of character loss caused by video image quality, complex background interference, OCR (Optical Character Recognition) limitation, dynamic subtitle change and the like. According to the method, key technologies such as self-adaptive subtitle denoising, semantic compensation and multi-frame fusion are adopted, and the completeness and accuracy of subtitle extraction are remarkably improved. The scheme is suitable for various types of movies, television dramas, short videos, conference videos and the like. The core innovation of the method is that lost or wrong characters can be effectively detected, compensated and corrected by intelligently analyzing the context and integrating multi-frame information, so that the subtitle output quality is greatly improved, and the effect is remarkable particularly in a scene that a single character is easy to lose.

The invention relates to a method, a device and equipment for identifying ancient characters of a war bamboo slip, and a medium. The method comprises the following steps: acquiring a to-be-identified war bamboo slip image containing a plurality of ancient characters; flattening a local feature graph containing local detail information of ancient characters in a to-be-recognized war national bamboo slip image output by the OfficientNetV2 convolutional neural network into a sequence structure, and splicing the sequence structure with a feature sequence containing global semantic information and long-range dependency information of the ancient characters output by a Transform encoder in a channel dimension to form a joint feature tensor; layer normalization processing is carried out; and inputting the combined feature tensor subjected to layer normalization processing into a classification module of an ancient character recognition model so as to output a single character recognition result corresponding to a single ancient character or character sequence recognition results corresponding to a plurality of continuous ancient characters. Compared with the prior art, the method has better supplementing and repairing capacity when processing defective and broken strokes, and the recognition capacity for characters with incomplete structures or complex forms is remarkably improved.

The invention relates to a character detection method and device, equipment and a medium. The method is applied to electronic equipment and comprises the steps that a to-be-processed image is acquired, the to-be-processed image comprises a character combination area, and the character combination area comprises a plurality of to-be-recognized characters; detecting single character regions corresponding to the plurality of characters from a character combination region contained in the to-be-processed image; segmenting each single character region in the character combination region to obtain characters in each single character region; a pre-trained character classification network is utilized to classify the characters in each single character area, the character category of each single character area is determined, the pre-trained character classification network comprises a first number of classification sub-networks, and the language category to which the characters belong comprises a second number of character categories; the first number is equal to the second number. Therefore, while the character detection cost is reduced, the accuracy and efficiency of character detection are improved.

The invention provides a composition correction method and device based on a large model, and the method comprises the steps: collecting a paper document containing a handwriting composition of a user, and generating a digital composition picture; performing stroke feature vector extraction on each single character; obtaining an identification result and confidence corresponding to each single character image; screening out single characters of which the confidence coefficient is greater than a first threshold value as quasi-recognition characters; determining the grammar accuracy of the statement where each quasi-recognition character is located; constructing a handwritten stroke library of the user according to the extracted stroke feature vectors; the stroke feature vector of each single character is corrected, and the recognition result and the confidence coefficient corresponding to each single character image are obtained again until the proportion of the number of the quasi-recognition characters is not lower than a second threshold value; determining composition content according to all single character recognition results; and based on a plurality of pre-trained composition evaluation large models, evaluating the composition content from different evaluation angles, and giving scores and comments. The composition correcting efficiency, accuracy and consistency are improved.

The invention relates to a character recognition system and method and a storage medium, and relates to the technical field of character recognition, and the system comprises an input module which is used for reading an original image; the preprocessing module is used for preprocessing the original image; the character detection module is used for extracting a character region image in the original image; the character segmentation module is used for segmenting the character region image to obtain a single character image; the character recognition module is used for extracting character features of the single-character image and inputting the character features into a character recognition model to obtain a character recognition result, and the character recognition model is used for determining a character probability value matched with the character features and determining the character recognition result; the decryption module is used for decrypting the encrypted characters; the output module is used for outputting the decrypted text; the control module is used for calling all the modules in sequence. According to the scheme, the time and the error rate of manual operation can be reduced, and the processing efficiency is remarkably improved.

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 invention discloses a speech recognition method and device, equipment and a storage medium. The method comprises the following steps: acquiring speech feature information of at least one speech chunk in speech to be recognized; time sequence classification processing is carried out on the speech feature information of a first chunk located at the initial position in the at least one chunk to obtain a first probability matrix of the first chunk, the first probability matrix comprises at least one probability vector, and the probability vector comprises probability values of multiple pinyin with tones; determining the phonetic alphabets with the probability values meeting a preset condition in the first probability matrix as target phonetic alphabets; and displaying a target single character corresponding to the target pinyin in a preset mapping table, wherein the preset mapping table comprises a plurality of pinyin with tones and a single character corresponding to each pinyin with tones. According to the embodiment of the invention, the first word on the screen is quickly determined, so that the perception of the user on the first word on the screen is improved, and the user experience of voice recognition is improved.

The invention discloses a network information content encryption method and system and a storage medium, and the method comprises the steps: obtaining network data in network traffic in real time, and carrying out the data preprocessing, and obtaining the preprocessed network information data; calculating a single character value according to the network information data and obtaining an encrypted character set; according to the network information data and the encryption character set, constructing an encryption matrix of the network information data to obtain the encryption matrix of the network information data; according to the network information data, performing data classification through a preset network information data importance sensitivity table to obtain network information classification data; performing data relation analysis according to the encryption matrix and the network information classification data to obtain an encryption importance matrix; and performing data encryption and transmission according to the network information data and the encryption importance matrix. According to the method, flexible classification encryption can be carried out according to the importance and sensitivity of the network information data.

The application provides a steel pipe multi-line handwritten character detection method and device, electronic equipment and storage medium, relates to the technical field of steel pipe handwritten character detection method, and comprises the following steps: extracting a multi-line handwritten character region on a steel pipe image to obtain a character region image containing multi-line handwritten characters; converting the character region image into a gray scale image; equally cutting the gray scale image along the vertical direction to obtain a plurality of vertical segment images; obtaining a corresponding projection curve for each vertical segment image by using a horizontal projection method, and determining a character line region of each line of characters according to the projection curve of the vertical segment image; extracting a single character region in the character line region; and performing character recognition on the single character region on the steel pipe image by using a preset recognition model to obtain a recognition result. The application can be applied to the recognition of handwritten characters on a steel pipe in an actual production process, and can efficiently and accurately recognize handwritten characters on the steel pipe.

The present invention provides an image processing device, an image processing system, an output device, an image processing method, and an image processing program that can improve the accuracy of character recognition for characters that are close to each other. [Solution] The acquisition processing unit 111 acquires image data containing characters. The rectangle candidate detection processing unit 113 detects character rectangle candidates from the image data and calculates the confidence level of the detected rectangle candidates. The single character extraction processing unit 114 corrects the confidence level of the rectangle candidates to be corrected among the first and second rectangle candidates, using a method for calculating the degree of overlap, which is selected according to the relationship between the type of the first character corresponding to the first rectangle candidate and the type of the second character corresponding to the second rectangle candidate, when at least a portion of the first and second rectangle candidates detected by the rectangle candidate detection processing unit 113 overlap with each other. The single character extraction processing unit 114 extracts single character rectangles based on the corrected confidence level.

Embodiments of the present application provide a speech recognition content correction method and device, electronic equipment and computer readable storage medium, first continuously obtain audio data of a unit time length, and an encoding feature sequence corresponding to the audio data; according to the encoding feature sequence, obtain a plurality of candidate recognition sequences, sequence scores of the sequences, and a plurality of time sequences corresponding to the plurality of candidate recognition sequences; in the case of receiving complete audio data, re-score the candidate recognition sequences to generate a target candidate recognition sequence; finally, according to the paragraph weight ratio of the time node where the character is updated, correct the character in the target candidate recognition sequence, and output the corrected target candidate recognition sequence as the final recognition result; by introducing paragraph weight analysis of the content relationship between a single character in the sequence and the context, and correcting individual characters one by one, the final output recognition text content is more accurate.

This invention provides a character recognition method, electronic device, and storage medium for a treatment control system display. The character recognition method includes: acquiring an image from the treatment control system display; acquiring an image to be recognized containing characters; binarizing the image to be recognized; removing edge blank areas from the binarized image; performing character segmentation on the binarized image to obtain several first character image blocks, where each first character image block is either a single-character image block or a connected character image block; confirming the category of the first character image blocks and performing the following processing: if the first character image block is a single-character image block, then by matching it with a template character image, the single character in the single-character image block is recognized; if the first character image block is a connected character image block, then the connected character image block is de-connected to recognize each character in the connected character image block. This invention enables automatic recognition of planning information in a treatment control system display.

The embodiments of the present disclosure disclose a method and device for character positioning. A specific implementation of the method includes: acquiring a target image; performing text line detection on the target image to obtain text line position information of the text line in the target image; performing character recognition on the image area in the target image located at the position indicated by the text line position information to obtain a character sequence included in the text line; and generating character position information of the characters in the text line based on the character sequence. This implementation determines the position of characters through text line detection and character recognition, which can improve the speed of single character positioning, reduce the amount of calculation for single character positioning, and reduce the consumption of computing resources during the positioning process.

This application provides a text-image matching method, an electronic device, and a storage medium, relating to the field of data processing technology. In this method, text is first acquired; then, the text is segmented to obtain segmentation results including single characters and phrases, where a phrase includes at least two single characters; next, the feature vectors corresponding to single characters are determined using the first embedding module of a pre-trained text encoder, and the feature vectors corresponding to phrases are determined using the second embedding module of the same text encoder; the first and second embedding modules are trained at different stages; subsequently, the feature vectors corresponding to single characters and phrases are encoded to determine the text semantic vectors corresponding to the text; finally, text-image matching is performed based on the text semantic vectors. This improves the accuracy of text-image matching and effectively enhances the user experience.

Input system (1) for inputting a code comprising a string containing at least numbers, wherein the input system (1) comprises: - a touch panel (2) configured to generate a contact detection signal (20) in response to the detection of a touch input gesture performed by a user (U) on the touch panel (2), wherein the contact detection signal (20) is an indicator of contact of one or more fingers (10) on the touch panel (2) during the touch input gesture; - a processing unit (3) configured to detect the contact detection signal (20) generated by the touch panel (2) and to select a single character depending on the contact detection signal (20) corresponding to the detected touch input gesture;wherein the input system (1) is characterized in that the processing unit (3) is configured to operate in an input mode to select numbers from a first string of characters comprising all or some of the numbers “0” to “9” by performing the following steps: - Determining that the touch input gesture is performed in the form of a main swipe movement (4) corresponding to the swipe movement of at least one finger (10) on the touch field (2) from a main starting point (41) to a main end point (42), where the main swipe movement (4) is interrupted by retracting the at least one finger (10), depending on the contact detection signal (20); - Determining a swipe movement vector (40) of the main swipe movement (4) depending on the contact detection signal (20);- Selecting a number from the first row of characters depending on the swipe movement vector (40) of the main swipe movement (4) independently of the main starting point (41) on the touch field (2).;

The invention provides an instant messaging method for automatically detecting sensitive information hiding, which comprises the following steps of: 1, aiming at a configured sensitive word set, constructing a sensitive word character sequence logic organization structure, and orderly bearing a single sensitive word in the sensitive word character sequence logic organization structure through a plurality of character mapping units on the same logic path, each character mapping unit corresponds to a single character in the sensitive word, the character mapping units pass through a preset logic association relationship, and the logic association relationship represents the sequence of the characters in the sensitive word; 2, on the basis of the constructed special character set, special characters in a to-be-filtered text generated in instant messaging are filtered out, and only non-special characters are reserved to form a to-be-matched character sequence; and step 3, matching the to-be-matched character sequence with the sensitive word character sequence logic organization structure to determine sensitive information in the to-be-filtered text. According to the scheme, the missed judgment problem of a traditional scheme is solved or relieved in a targeted manner.

The invention relates to a multi-line text searching and replacing method and system, and the method comprises the steps: carrying out the preprocessing of an original multi-line text, a multi-line searching content and a multi-line replacing content, removing the head and tail spaces of each line, and obtaining a corresponding logic text, a logic searching content and a logic replacing content, judging whether the logic search content contains a preset wildcard character or not, if the logic search content does not contain the preset wildcard character, carrying out simple character string matching on the logic text to obtain a logic index search result, converting the logic index search result into an original character index by utilizing a preset index mapping relation for positioning, and executing simple replacement operation, and if the logic search content contains the preset wildcard character, carrying out simple character string matching on the logic text to obtain a logic index search result; and if the logic text is matched with the fuzzy item list, regarding the logic search content as an integral matching block, performing sliding comparison according to rows in a row array corresponding to the logic text, if the matching succeeds, extracting the fuzzy item list and the starting and stopping row indexes of the matching block, performing positioning by utilizing an index mapping relation, and executing fuzzy replacement operation of inserting the fuzzy item list into a placeholder.

The present invention discloses a voice-interactive teaching system that integrates a visual-tactile sensor and a Braille recognition algorithm. The system comprises: obtaining a Braille image through a tactile sensing module, obtaining an input instruction through a voice module, and adjusting a mode of a Braille translation module, wherein the adjusted modes include a teaching mode and a translation mode; in the teaching mode, the Braille image is preprocessed by the Braille translation module to obtain dot information, the dot information is recognized by the Braille recognition algorithm to obtain character information, and the character information is broadcasted through the voice module; in the process of recognizing single character information, the use of the Braille recognition algorithm also includes determining whether to use the Braille recognition algorithm for detection through a frame difference algorithm and an inter-frame minimum center distance algorithm; in the translation mode, the character information is detected through four columns of threshold lines according to the dot information to obtain the detected character information, and the detected character information is broadcasted through the voice module.

The invention discloses a character detection method, medium and system based on optical character recognition, and the method comprises the steps: obtaining a screen picture, and carrying out the binarization processing of the screen picture, so as to generate a binary picture corresponding to the screen picture; performing single character recognition on the binarized picture to obtain frame coordinate information and confidence coefficient corresponding to each character in the binarized picture; filtering each character based on a preset confidence threshold; for each filtered character, judging whether each character is an abnormal character or not based on the frame coordinate information; if yes, marking the abnormal characters; the abnormal characters in the text can be effectively recognized, and the situation that the abnormal characters are deleted by mistake and cannot be read is prevented.

The invention provides a document analysis method and device, equipment, a storage medium and a product, and the method comprises the steps: carrying out character prediction based on a document page image and text prompt information indicating analysis, and generating a target character; performing character prediction based on the target character and a set prediction guide unit, and generating at least two characters of which the character positions are ranked behind the target character; if the ending character is not obtained, taking at least two characters as target characters, and performing character prediction again based on the target characters and the prediction guide unit until the ending character is obtained; and based on the characters obtained by each prediction, an analysis result is generated, and the analysis result comprises text information used for representing the content elements in the document page image and the structural relationship between the content elements. According to the technical scheme, the prediction guiding unit is introduced, so that multiple characters can be predicted in parallel in single character prediction, and the document analysis speed is increased while the document analysis accuracy is kept.

The invention discloses a method and device for recognizing and reconstructing vector characters of a CAD drawing, and relates to the technical field of data processing. The method comprises the steps that a PDF file is analyzed, and vector elements and attribute information of the vector elements are extracted; character stroke candidate lines are screened out according to attribute grouping; carrying out connectivity analysis and density-based spatial clustering on the lines, and identifying a text region; rendering the vector lines of the text region into an image with high resolution, carrying out single character segmentation, and recognizing characters by combining drawing-level font library template matching with a single character recognition model; combining the single characters into word groups according to the positions and the context information, and performing word order correction by adopting a language model; the recognition result is written back into an editable CAD text object; according to the method, vector geometric information is directly utilized, distortion of bitmap sampling is avoided, the small character recognition rate and the anti-jamming capability are effectively improved, and end-to-end semantic recovery from vector lines to searchable and editable texts is achieved.

The invention provides a report writing content-based wrongly written character detection reminding method, which comprises the following steps of: obtaining a preliminary report statement needing to be detected, if characters and / or words in the preliminary report statement are characters and / or words in a blacklist medical wrongly written character database, judging that the characters and / or words are wrongly written characters and defining as first wrongly written characters, and at the moment, judging that the characters and / or words are wrongly written characters and defining as second wrongly written characters; extracting correct content corresponding to the first wrongly written character from the blacklist medical wrongly written character database, and replacing the first wrongly written character with the correct content; based on a white list NLP word library, performing white list word segmentation on the modified preliminary report statement, and if a single character which cannot be divided appears, determining that the single character is a wrongly written character, and defining the single character as a second wrongly written character; and extracting the word with the highest probability from the white list NLP word bank according to the pinyin as the modification content, and returning the modification content to the report. The invention further discloses a system for detecting and reminding wrongly written characters based on the written content of the report. According to the method, the wrong content of the report text is recognized, a doctor is reminded to correct wrongly written characters, and the problems of false report and missing report of the wrongly written characters in the text report are avoided.

The invention provides a character practice method and device, electronic equipment, a storage medium and a product. The method comprises the following steps: shooting in real time to obtain a user exercise book image; identifying grids in the exercise book image and determining a to-be-exercised area; the to-be-practiced area is matched with a standard grid in a character action model to determine the display position of the virtual action model in the to-be-practiced area, and the character action model is a dynamic three-dimensional model containing a hand action track in the process that a single character is written in the standard grid; the to-be-exercised characters are exercised based on the display position and the character action model, a learner can learn according to own time arrangement and rhythm and can also select characters needing to be exercised according to own requirements, the space-time flexibility of character exercise is improved, and personalized teaching is achieved.

The invention discloses a handwritten original record recognition method based on an OCR (Optical Character Recognition) technology, which relates to the technical field of OCR recognition, and comprises the following steps: setting a record recognition range, and collecting an image of handwritten original record information based on the record recognition range; preprocessing the image of the handwritten original record information to obtain initial data information; a character loader is set, initial data information is input into the character loader, error correction points are set based on the character loader, and the error correction points comprise a migration error correction port and a stroke comparison database. According to the method, the error correction points including the walk error correction ports and the stroke comparison database are arranged, in the recognition process, the walk error correction ports can perform preliminary feature extraction on single characters, local feature recognition is performed on the characters through the recognition antennas, detail character features and preliminary character features are combined into identifiable characters, and the recognition accuracy is improved. And the character feature description is more complete.

The application discloses an image recognition method and device based on chroma domain positioning and topological constraint, relates to the technical field of data processing, and comprises the following steps: after an input image is preprocessed, an initial mask is generated in combination with contrast criteria and darkness criteria, pseudo targets are removed through hole-preserving connected domain analysis and Euler number filtering; rendering is performed by using an artificial color embedding technology, chroma domain positioning is performed in an HSV space, and an MRZ region is extracted; an inclination angle is estimated based on edge line direction consistency, and a maximum inscribed rectangle cutting is performed to obtain an MRZ cutting region; character segmentation is performed based on format constraint, an uncertainty conduction scheduling mechanism is adopted to adjust segmentation parameters and a retry strategy, and multiple single-character images are output; HOG features are extracted from the single-character images, a character classification is performed by using an SVM classifier, and preliminary recognition results and classification scores of each character are obtained; a final recognition score is fused based on structure prior scores and classification scores, and complete MRZ text is output.

The invention provides an automatic limit line forming circular grid writing method, relates to the technical field of Chinese character writing training of primary and secondary school students, and aims to solve the core pain points that the font size is difficult to control during writing of the primary and secondary school students and the handwriting is easy to skew when being switched to a transverse line / blank paper. The core of the method is a round grid composed of an outer circle with the diameter of 6 mm and a central inner square with the side length of 2 mm, the round grid is matched with a standard composition grid with the diameter of 8 mm, and a blank space with the diameter of 1 mm is formed; the method is realized through four-step training of practicing a single character by a single round grid, practicing coherent writing by a row of round grids, alternately transitioning round grids and transverse lines, and challenging a reference line at the top of blank paper, and is combined with a special exercise book, an indentation ruler or light-color printing. By means of the method, the paper surface neatness of composition grid writing can be immediately improved, the students can form muscle memory through training of the psychological datum line, finally, the students can also write neatly on the blank paper, and the method is suitable for the students of the third grade and above of the primary school and is a scientific and practical writing training system.

The present application relates to the image processing technical field in machine vision, especially relates to the high-voltage transmission line inspection image processing field in power system, and specifically relates to a kind of character detection and identification method in power pole information board image.The power pole information board image is preprocessed first;Then the feature points of the power pole information board image after preprocessing are obtained, and the character candidate region is obtained;Remove the non-character region in the character candidate region, obtain single character region;Single character region is merged into text line;Text line is recognized by character.This application makes full use of the advantages of MSER detection natural scene text, gives full play to the advantages of Imbinarize-MSER local feature multi-scale detection, candidate region stability is good, affine invariance, and the character detection and identification of the power pole information board collected, facilitate monitoring and management the running state of transmission line, provide security for the safety of transmission line.

A training method includes: obtaining a synthesis text image set, a synthesis text image being obtained through synthesizing a real scenario background image and a random word, the synthesis text image being provided with a line text annotation box and a single character annotation box; training an initial algorithm network using the synthesis text image set, so as to obtain an intermediate model; processing a real scenario text image set using the intermediate model, so as to obtain a pseudo label of a real scenario text image, the real scenario text image being provided with a line text annotation box, the pseudo label being a single character annotation box; and training the intermediate model using the synthesis text image set and the real scenario text image set having the pseudo label, so as to obtain the single character detection model.

The application discloses an intelligent display voice-to-text method and system and a medium, and the method comprises the following steps: when an audio and video file is captured, the audio and video file is converted into text; each sentence of text is segmented according to a pre-agreed punctuation mark to obtain segmented sentence text information; a pre-established dictionary table is searched; if historical data corresponding to the segmented sentence text information is found in the dictionary table, a corresponding previously combined text paragraph is obtained from the dictionary table; the text paragraph is rendered; single characters in the rendered text paragraph content are processed according to sensitive words, forbidden words or search words; and the processed single characters are synchronized with the text paragraph on a page for display. Through the application, a user can quickly find out where a violation is located, and can also quickly jump to a corresponding progress for manual auditing to confirm whether a problem exists, thereby greatly reducing the workload of the user for compliance processing.

The invention discloses a form information extraction method and system, and the method comprises the steps: obtaining a form image, carrying out the character recognition of all character pre-selection boxes on the form image through a preset text recognition technology, obtaining a character recognition result, building a first image according to the character recognition result, carrying out the histogram comparison of the first image and a second image, and obtaining a form information extraction result. The second image is a region image of the whole text region in the form image, if the histogram comparison result meets the requirements of a filter, generating target text data, and if the histogram comparison result does not meet the requirements of the filter, generating target text data; and extracting an image region from all the character pre-selection boxes on the form image, and performing single character detection to output target text data.