851 results about "Chinese characters" patented technology

The invention discloses an archive data management labeling method and device based on computer application, and relates to the technical field of archive information management and intelligent labeling, and the device comprises a semantic analysis module, an allogenic character normalization module, a context modeling module, a historical semantic alignment module, a context feedback module, a processing module and a model optimization module. The device can automatically monitor file input, clean and segment texts, quickly and normatively replace foreign characters and intelligently convert simple and traditional Chinese characters; a semantic vector is generated based on a Transform model, semantic consistency is detected, and if semantic drift is found, rollback adjustment is carried out; the processing module is combined with a rule engine and a depth model to complete entity labeling and multi-label reasoning; and the optimization module dynamically records and feeds back annotation data, and supports self-learning and continuous improvement of the model. By continuously optimizing the judgment logic, the system has self-learning and evolutionary capabilities, so that the archive labeling process is more and more intelligent, and the labeling result is more and more accurate.

The invention discloses an adaptive rendering method and system for characters and images in AI digital human virtual and real scene fusion, and the method comprises the following steps: S1, collecting Chinese text and image data, and extracting semantic and regional features; s2, constructing a Shenchang differential equation converter model, and inputting text and image features to generate fusion representation; s3, extracting an attention weight matrix, and determining the spatial arrangement position of the Chinese text in the image; s4, optimizing the structural parameter vector by adopting a luminous insect colony optimization algorithm to obtain a fusion model; s5, generating an image-text rendering instruction set based on the preliminary arrangement and optimization model; s6, in combination with the digital human action state and background image information, attitude mapping and illumination adjustment are completed to generate a rendering frame; and S7, inputting the rendering frame into a rendering engine, and outputting the video content fused by the Chinese characters and the images. According to the method, accurate fusion of Chinese semantics and images is realized, and the naturalness and interactivity of AI digital human image-text presentation are remarkably improved.

The invention relates to a Chinese language and literature computer networking query reading system. According to the system, a query request processing module carries out mixed feature analysis processing according to a query request of a user, and a mixed query vector fusing radicals and semantics is generated; a distributed retrieval module performs distributed retrieval in a preset space-time fragment index structure based on the mixed query vector to obtain a candidate literature data set; the preliminary retrieval optimization module is used for carrying out self-adaptive conversion processing on abnormal characters and traditional and simple characters in the candidate literature data set according to a Chinese character evolution rule and a context to obtain standardized text data; and the visual query output module performs knowledge graph enhancement processing on the standardized text data to generate a visual query result. According to the system, efficient query and visual display of Chinese language and literature works are achieved through multi-modal processing, multi-dimensional indexing and knowledge graph technologies, and the problems that a traditional query mode is single in input, limited in retrieval dimension, insufficient in text processing and the like are effectively solved.

The invention discloses a computer data storage method and system, and relates to the technical field of computer data storage.The computer data storage method comprises the steps that variant samples are generated through multi-modal feature extraction, font topology analysis and adversarial training, and fine modeling of Chinese character connection and pixel-level interference in an electric power drawing is achieved; an isolated storage pool path is dynamically generated through a tree-shaped namespace, and a physical storage address and version information can efficiently and uniquely correspond and retrieve in combination with bidirectional mapping of version identifier construction and block chain indexes; according to a confidence score obtained through weighted calculation of the topological similarity and the term library matching degree, once the confidence score is lower than a preset threshold value, the system pushes a manual recheck mark to a specified terminal through an asynchronous message queue, and meanwhile, read-only data locking is implemented to ensure that the writing process is not interrupted; the consensus management mechanism is based on an improved practical Byzantine fault-tolerant algorithm, and block chain index consistency and data traceability security are guaranteed.

The invention discloses a 4HPU-based natural language character and numerical value general solution system and method and application thereof, and aims to solve the technical problem of realizing unification, universality and reasonability of natural language character and numerical value codes. A main coding path for realizing coding generation and structure nesting is arranged; the method comprises the following steps of: constructing a basic coding unit, judging the types and codes of numerical values, letters, Chinese characters or symbols; compared with the prior art, the method has the advantages that natural numbers, letters, Chinese characters and symbols are subjected to structured coding in a right-to-left nesting mode, complete coded word strings are obtained, structural expression and path reasoning are clear, dynamic expansion and compression can be achieved, and the coding efficiency is improved. And a more excellent basic coding method is provided for application scenes of data recovery and reverse recognition, natural language processing, intelligent communication and low-power-consumption equipment.

The invention relates to the technical field of oracle paraphrasing, and provides a training method and device for an oracle single-character image reasoning task. According to the method, a backbone network is trained to perform Chinese character recognition, modern Chinese character radical splitting, literal period identification and recognition, oracle-bone image recognition, oracle-bone multi-graph statement recognition, oracle-bone character meaning evolution analysis and font thinking chain reasoning on single-character images; calculating a joint loss value according to the primary output, the intermediate output and the advanced output obtained by training so as to iteratively update network parameters of the backbone network, and obtaining an intermediate model until the joint loss value meets a training ending condition; and training the intermediate model to perform font thinking chain reasoning, and finely adjusting the intermediate model by using a reward function to obtain a target model, thereby solving the problem that the oracle single-character image is difficult to accurately identify, infer and explain.

The invention discloses a Chinese character structure and geometric space interactive writing method, which relates to the technical field of interactive writing methods, and comprises the following steps: carrying out geometric classification processing on Chinese characters by adopting a structural analysis method to obtain a Chinese character type set comprising a left-right structure, an up-down structure and a surrounding structure; carrying out geometric rule definition processing on the obtained Chinese character type set to obtain a geometric constraint rule base containing component proportion, gravity center alignment and stroke topological relation; performing structured storage processing on the obtained geometric constraint rule base by adopting a database construction method to obtain a dynamically-called Chinese character writing template database; the writing process of the user is monitored in real time, and a writing feature data set containing stroke coordinates, writing speed and pressure data is obtained; performing matching processing on the obtained writing feature data by adopting a dynamic grid generation method to obtain a self-adaptive reference grid corresponding to the currently written Chinese character; and carrying out deviation detection processing on the obtained self-adaptive reference grid by adopting a computer vision method.

The invention provides an external plug-in system based on retrieval enhancement. The external plug-in system comprises an indexing module, a retrieval module and a generation module, the index module is used for performing cleaning, segmentation, organization, semantic coding and storage tasks of text data; the index module comprises a dynamic data management sub-module, a strategy decision case library management sub-module, a Chinese character recursive segmentation sub-module, a tree index structure construction sub-module and a vectorization storage sub-module; the retrieval module is used for completing semantic matching and sorting on the basis of the structured index; the retrieval module comprises a double-tower rough arrangement sub-module, a cross fine arrangement sub-module, a real-time update sensing sub-module and a cache management sub-module; the generation module is used for carrying out integration and semantic modeling on the high-quality intelligence information provided by the retrieval module; the generation module comprises an information integration and reasoning sub-module and a credibility evaluation and feedback sub-module; the method supports fact alignment, real-time updating, logic interpretability and output evaluability.

The application relates to a combined device and method for vertical static load test of water anchor pile method. Two gantries are arranged at the two ends of a water bearing platform, a steel beam passes through the space of a to-be-tested foundation pile connecting assembly I and an anchor pile connecting assembly II in the shape of a Chinese character 'jing', and the two ends are respectively suspended below the cross beams of the gantries; the to-be-tested foundation pile connecting assembly I is arranged in the middle of the steel beam, two anchor pile connecting disc assemblies II are arranged at the two ends of the steel beam, the lower end surface of a connecting disc I of the to-be-tested foundation pile connecting assembly I and the lower end surface of a connecting disc II of the anchor pile connecting assembly II are located on the same horizontal plane; a pulling jack is arranged between an upper tensile small cross beam I of the to-be-tested foundation pile connecting assembly I and the steel beam; a compression jack is arranged between the steel beam and the connecting disc I of the to-be-tested foundation pile connecting assembly I; and a recording platform is arranged beside the to-be-tested foundation pile below the middle part of the steel beam. The application combines the compression and pulling static load test devices into one, can be easily converted, is quickly installed, and the test result is reliable.

The invention discloses an AI automatic editing method and related equipment, and the method comprises the steps: obtaining a video prototype template selected by a user, prompt information inputted by the user, and original materials uploaded by the user through a user interaction interface, and calling a language large model to generate a structured video script; performing association degree matching on the structured script and the original material through a target detection model, and screening the original material according to a matching result and a quality scoring model to obtain an original video clip; performing background music matching according to the video script to obtain background music, generating voice dubbing and a flower-character chartlet according to the video script, generating video subtitles according to the voice dubbing, performing time axis alignment on each video media element, and synthesizing into a finished video. According to the embodiment of the invention, full-process automatic processing from script generation to video synthesis can be realized, and a high-quality finished video meeting user requirements can be efficiently output. The method can be widely applied to the technical field of automatic editing.

The invention discloses a multi-layer concrete structure construction method based on an embedded support shaped like a Chinese character'tu ', belongs to the technical field of building construction, and solves the problems of overall displacement and out-of-control precision caused by the fact that the bottom of a supporting system is not firmly fixed in traditional multi-layer concrete construction. According to the technical scheme, the method comprises the steps of firstly manufacturing a steel bar welded support shaped like a Chinese character'tu '; pre-burying the support system in lower-layer concrete to serve as a bottom fixing point of a vertical rod of the support system; a supporting system vertical rod is detachably connected with the top of the support; concrete is poured after the formwork is installed and adjusted; cutting the exposed part of the bracket after the concrete is hardened; and repeating the steps to finish multi-layer construction. According to the method, through the rigid pre-embedding and detachable connection design of the support shaped like the Chinese character'tu ', the stability and positioning precision of a supporting system are ensured, and the method is mainly used for rapid and high-precision construction of cast-in-place concrete structures such as high-rise buildings or multi-layer industrial factory buildings.

The invention discloses a Chinese character image super-resolution reconstruction method and system based on a diffusion model, and the method comprises the following steps: S1, constructing a Chinese steganography image data set, and generating a high-resolution image and low-resolution image pair; s2, adopting a conditional denoising diffusion model as a framework, taking the Chinese steganography image data set constructed in the step S1 as conditional input of a diffusion model, and guiding the diffusion model to learn mapping from low resolution to high resolution; and S3, performing resolution reconstruction on a to-be-processed low-resolution Chinese character image by using the conditional denoising diffusion model in the step S2, and outputting a high-resolution Chinese character image. According to the method, the problems of fuzzy details, insufficient generalization ability and insufficient semantic information utilization in Chinese character image super-resolution reconstruction in the prior art are solved.

The invention relates to a Chinese character style migration method and a migration system thereof. The Chinese character style migration method comprises the following steps: acquiring Chinese character image data to be subjected to style migration; inputting the Chinese character image data into the trained neural network model; obtaining Chinese character image data after style migration; the neural network model comprises a generator module, a discriminator module, a mapping network module and a StyleEncoder style encoder module, the generator module extracts spatial features based on Chinese character image data and normalizes the spatial features, required style picture data is generated in combination with the StyleEncoder style encoder module, and the StyleEncoder style encoder module is used for identifying the Chinese character image data. The discriminator module discriminates the authenticity of the picture data; according to the Chinese character style migration method and the migration system thereof, the feature acquisition capability of the model is enhanced, and the accuracy of the generated picture style is improved.

The invention relates to the technical field of image detection, can be applied to business system platforms of financial science and technology, medical health and the like, and discloses a Chinese character font generation method, device, equipment and medium, the method comprises the following steps: obtaining a Chinese character image to be processed, and carrying out image coding on the Chinese character image to be processed to obtain a Chinese character image sequence; performing style feature extraction and semantic feature extraction on each Chinese character image vector in the Chinese character image sequence to obtain corresponding Chinese character style features and Chinese character semantic features; constructing a corresponding style loss function and a semantic loss function according to the Chinese character style features and the Chinese character semantic features; performing loss weighting calculation on the style loss function and the semantic loss function to obtain a target loss function; and performing font style generation on the to-be-processed Chinese character image according to the target loss function to obtain a target Chinese character font of the to-be-processed Chinese character image. The Chinese character font generation efficiency and font generation diversity can be improved.

The invention discloses a planar two-dimensional micro-displacement sensor based on a four-quadrant eddy current effect, which comprises a detection coil array plate and a target conductor array plate which are parallel up and down, and is characterized in that the detection coil array plate is provided with four planar coils arranged in a shape like a Chinese character'tian 'on an insulating substrate; the target conductor array plate is provided with four metal panels which are arranged in a shape like a Chinese character'tian 'on an insulating substrate. According to the invention, identical excitation signals are introduced into four planar coils on a detection coil array plate, two-dimensional micro-displacement of the detection coil array plate relative to a target conductor array plate is converted into voltage signals of the four planar coils by using an inductance measurement circuit, and the four voltage signals are processed by using a microprocessor system; therefore, the two-dimensional plane relative displacement of the detection coil array plate and the target conductor array plate is obtained. The device is simple in structure, high in measurement precision and suitable for scenes requiring precise measurement of planar two-dimensional micro-displacement in severe environments such as dust and greasy dirt.

The invention relates to a controllable Chinese character new font generation method and system based on a diffusion model, and belongs to the technical field of computer image processing. Based on the system, a user inputs a character image of a target stroke style and a character image of a target skeleton, then multi-scale skeleton features are extracted through a skeleton information extraction network, stroke style features are extracted through a stroke style extraction network, and the multi-scale skeleton features are extracted. And finally, denoising operation prediction of each step of the U-net network in the diffusion process is guided through the features, so that a new font character image conforming to expectation is generated. The method is inspired by a structure reference method, features are extracted from the input image through the neural network, a diffusion model is guided through skeleton features and stroke features to generate a new font style, and basic requirements of users are met.

This invention discloses an automatic robot instruction inspection method and apparatus based on the PDPS platform. The method involves acquiring offline program data of the robot and the corresponding design rules, configuring a program inspection tool according to the design rules, inputting the program data into the inspection tool, and storing safety information from the first and second instructions into corresponding variables. The information in each variable is compared or judged to obtain a first judgment result. Chinese characters and punctuation marks in the program data are judged to obtain a second judgment result. Based on the motion parameters in the program data, the margin value of each joint when the robot performs each action is calculated, and collision risk is judged based on the margin value to obtain a third judgment result. Finally, the robot's compliance with safety rules is judged based on the motion radius to obtain a fourth judgment result. This method improves the accuracy of offline program inspection.

The invention discloses a Chinese lip language recognition method based on credible visual position element acquisition. The method comprises the following steps: S1, data acquisition and preprocessing: obtaining video data depicting lip motion; s2, deep clustering: performing deep clustering on the lip motion depicting video data to obtain the number of cluster-distributed visual position classes, corresponding visual position classes and a visual position library, so as to obtain frame-by-frame image data with visual position class labels corresponding to the lip motion depicting video data; and S3, identifying the cascade Chinese character sequence based on the visual position element intermediate representation: performing feature extraction based on the frame-by-frame image data with the visual position element category label to realize the identification of the cascade Chinese character sequence taking the visual position element as the intermediate representation. According to the method, the cumulative error of recognition prediction can be reduced, the lip language recognition performance based on the visual position elements is improved, and the accuracy bottleneck of the lip language recognition based on the visual position elements is broken through.

The invention relates to a forging method of an aluminum alloy rectangular-ambulatory-plane frame, which comprises the following steps: selecting a cast ingot or a bar billet as a raw material, performing three-way upsetting and pulling modification to prepare a square transition billet with a square plate and a concave step, and positioning the concave step between the two ends of the square plate in the length direction; a flat anvil is used for conducting multi-pass forging on the inwards-concave steps, the two ends of the square plate are widened to the matched size after each pass of forging is completed, the square transition blank is flattened to form a square plate blank, and inwards-concave grooves are formed in the center positions of the plate faces of the two sides of the square plate blank; a pre-forging die is used for pre-forging the square plate blank, so that the square plate blank is shaped into a frame type pre-forged piece with a wad in the center; and a finish forging die is used for conducting finish forging on the frame type pre-forged piece, so that the frame type pre-forged piece is forged into a frame finish forged piece shaped like a Chinese character'hui '. According to the method, concave grooves are formed in the center positions of the two side plate surfaces of the square plate blank during pre-forging, so that the square plate blank can be formed into a hollow-square-shaped frame final forging piece only through one-time pre-forging shaping and final forging.

The invention relates to the technical field of natural language processing, in particular to a data enhancement and feature enhancement-based Chinese named entity recognition method, which is based on a feature enhancement double-attention named entity recognition model, and is characterized in that the model comprises an embedded layer, a data enhancement module, a multi-scale convolution fusion attention layer and a prediction layer; the embedded layer uses a dual-channel attention fusion module to process texts in parallel, fuses multi-dimensional information features of Chinese characters, combines an error correction type mask language model, a pre-training model and a bidirectional gating loop unit, fuses local and global text features, and obtains text representation from multiple dimensions and multiple levels; according to the method, by introducing an innovative model mechanism or training strategy, challenges such as label sparsity and text noise existing in a Chinese named entity recognition task can be effectively handled, the learning ability and generalization performance of the model for long-tail entity categories are improved, and the robustness of the model in a real and non-ideal data environment is enhanced.

The invention provides a water level data anomaly detection processing method based on multi-method fusion. The water level data anomaly detection processing method comprises the following steps of data acquisition and environment configuration; library importing and environment setting: importing a pandas library, a numpy library and a matplotlib library for data processing and visualization, realizing an isolated forest algorithm by sklearn.ensemble. Isolation Forest, and drawing a box graph by means of seaborn; setting Chinese font display: ensuring normal Chinese display of the chart; and collecting and classifying multi-source data. According to the water level data anomaly detection processing method based on multi-method fusion provided by the invention, the provided water level data anomaly detection processing method based on the fusion of the box graph model, the isolated forest model and the median filtering + 3sigma model is utilized to detect and clean the water level data of the selected gate station in one month, and the RMSE and the MAE are calculated to obtain the abnormal water level data of the selected gate station. A detection method suitable for three conditions of high-precision requirement, high-risk scene comprehensive screening and rapid analysis is obtained; drawing a curve graph of the original data and the processed data, marking abnormal points, and visually comparing the cleaning effects of the three methods.

The utility model belongs to the technical field of accounting document gluing, and particularly relates to a portable accounting document gluing device which comprises a gluing platform and a gluing mechanism and further comprises an n-shaped frame and a pressing mechanism, and the gluing mechanism comprises a moving table shaped like a Chinese character'hui ', a glue storage barrel and a gluing wheel; the pressing mechanism is installed on the n-shaped frame and used for pressing the accounting documents downwards. The moving table shaped like the Chinese character'hui 'is arranged on the n-shaped frame and has the moving freedom degree in the horizontal direction, the glue storage barrel penetrates through the moving table shaped like the Chinese character'hui' and has the moving freedom degree in the vertical direction, a glue outlet part is fixed to the bottom end of the glue storage barrel, and a glue outlet is formed in the bottom end of the glue outlet part. The gluing wheel is rotationally arranged in the glue outlet; the accounting documents can be pressed downwards through the arranged pressing mechanism, the stability of the accounting documents is guaranteed, manual pressing is not needed, and gluing is more convenient; and the arranged gluing mechanism moves along the guide sliding hole to conduct gluing, it can be guaranteed that gluing is conducted along a straight line, and the gluing accuracy is improved.

The invention relates to the technical field of liquid crystal display, and discloses a liquid crystal module character display method, which comprises the following steps of: performing stroke decomposition and topological relation establishment on Chinese characters in a target character set to generate a character stroke data set; performing two-dimensional character topological mapping and compressed encoding on the to-be-displayed characters to form compressed character description data; executing hierarchical rendering and stroke synthesis in a character reconstruction engine to generate a target character dot matrix; pixel array electric field distribution optimization is carried out, and sub-pixel-level precision character edge display data are obtained by controlling the driving voltage gradient; performing electric field compensation based on the character edge display data with sub-pixel-level precision to form character display data with enhanced edge sharpness; stroke self-adaptive anti-aliasing processing is carried out on the character display data with enhanced edge sharpness, clear Chinese character display data is output, the visual smoothness of character edges in a small-size display environment is optimized, and the overall recognition degree of characters is improved.

The invention provides a cross-font Chinese character recognition method based on structure embedding alignment, which aims at solving the problems of large stroke form change and strong structure variation of Chinese characters under different font styles, and comprises the following steps of: firstly, carrying out skeleton extraction on a character image and modeling into a topological graph structure with geometric attributes; extracting character graph embedding representation through a graph converter network comprising graph convolution, graph attention and a graph compression module; meanwhile, a structure tree corresponding to characters is constructed, a structure vector is generated by utilizing a structure embedding model, graph representation is supervised through structure consistency loss, and model learning is guided to have embedding features with structure stability and semantic consistency. According to the method, the recognition capability of the model for complex fonts, structural shielding and fuzzy characters is effectively improved, the structural understanding and generalization performance in a cross-font recognition task are remarkably enhanced, the problem that the recognition accuracy is reduced when font differences are large in an existing method is solved, and the method is suitable for various character image processing scenes.

The invention relates to the technical field of risk portrait generation based on a knowledge graph, and discloses a risk portrait generation method based on the knowledge graph. According to the scheme, the document is segmented through the sentence end symbols, semantic coherence is ensured, and the accuracy and stability of candidate word extraction are improved. Continuous Chinese character sequence segmentation is adopted, dependence of a word segmentation tool is avoided, the occurrence frequency and position of candidate words are recorded, and a detailed statistical basis is provided for risk factor screening. A minimum occurrence number constant is set, noise interference and information coverage are balanced, the screening flexibility is improved, and candidate word screening is more representative. And counting the co-occurrence relationship of candidate words in the sentence, quantifying the co-occurrence frequency through an indicator function, enhancing word association expression, and providing key data for initial graph construction. Finally, the scheme constructs a knowledge graph based on the candidate words and the co-occurrence relation thereof, ensures clear expression of risk information, and improves the accuracy of the system in risk identification and portrait generation.

The utility model belongs to the technical field of tourniquets, and particularly relates to a spinning tourniquet which comprises a self-adhesive tape, a buckle shaped like a Chinese character'ri 'is fixedly connected to one end of the self-adhesive tape, a screwing mechanism is arranged on the self-adhesive tape, and the screwing mechanism is close to the buckle shaped like the Chinese character'ri'. The screwing mechanism comprises a pressurizing assembly arranged on the self-adhesive tape, the pressurizing assembly is in transmission connection with a screwing assembly, a self-locking structure is arranged in the screwing assembly, a timer is arranged on the screwing assembly, and a marking pen is detachably connected to the screwing assembly. The spinning tourniquet is provided with the timer and the marking pen, a user can conveniently record the hemostasis time, and the situation that the hemostasis effect is poor due to the short hemostasis time or the limbs are damaged or necrotic due to the long hemostasis time is avoided.

The invention discloses a cross-modal environment interaction method and a cross-modal environment interaction system for fusing physical patterns and Chinese character elements. Equipment object images and Chinese character control instructions are dynamically synthesized into an interactive interface through a semantic association engine, natural intention recognition is achieved in combination with a multi-mode input decoding technology (touch control / voice / electroencephalogram and the like), and terminals such as a wall panel and a robot screen are adapted under a layered rendering architecture. According to the system, the interaction intuition and the culture affinity are remarkably improved, and the cognitive load problem of a traditional GUI in cross-equipment and cross-culture scenes is solved.

The utility model relates to the technical field of high-voltage mutual inductors, and discloses an improved high-voltage mutual inductor coil which comprises an iron core, the iron core is shaped like a Chinese character'hui ', a secondary winding coil is wound on the top edge of the iron core, a primary winding coil is wound on the outer circle wall face of the secondary winding coil, and a sponge strip is arranged on the outer wall face of the iron core. A fixing structure is arranged on the outer wall face of the iron core. According to the utility model, through the arrangement of the fixing structure, after a worker adheres a sponge strip to the outer wall surface of the iron core, the two fixing frames are close to each other to wrap the outer wall surface of the iron core, then the two supporting plates are connected through bolts, and the bolts are tightened to enable the left and right inner walls of the two fixing frames to abut against and fix the sponge strip; and then a worker rotates a bolt at the mounting plate, so that the bolt of the mounting plate abuts against the bottom surface of the fixing frame, and the movable plate abuts against and fixes a sponge strip on the top surface of the iron core, and in the mode, the durability of fixing the sponge strip is improved, and the use effect of the high-voltage mutual inductor coil is guaranteed.

1. The name of this design product: Lumbar support cushion (in the shape of the Chinese character "Fu"). 2. Purpose of this design product: This design product is a lumbar support cushion used to correct sitting posture. 3. The key point of the design of this product lies in its shape. 4. The picture or photo that best illustrates the design points: three-dimensional picture.

The invention discloses a Chinese character recognition method based on Chinese character similarity algorithm joint optimization. Comprising the steps of obtaining a to-be-recognized Chinese character image, inputting the preprocessed image into a convolutional neural network classifier, and outputting a Chinese character category prediction result; obtaining a corresponding standard Chinese character image from the Chinese character database, and respectively calculating stroke similarity, structural similarity and topological similarity between the to-be-recognized image and the standard image; calculating a Chinese character similarity score based on multiple similarity fusion; and constructing a joint loss function containing classification loss and similarity loss, and carrying out joint training on the classifier network. By introducing a stroke template feature extraction and dynamic time warping algorithm, accurate matching of stroke forms and sequences is realized. The technical problem that similar Chinese characters are difficult to distinguish in a traditional recognition method is effectively solved, deep fusion and collaborative optimization of classification and similarity information are achieved, the accuracy and stability of Chinese character recognition are remarkably improved, and the method is suitable for diversified application scenes.