21 results about "Character analysis" patented technology

The invention relates to the technical field of soybean plant information analysis, in particular to a soybean whole plant character analysis method based on three-dimensional point cloud. Comprising the following steps: carrying out collection, labeling and standardized pretreatment on three-dimensional point clouds by adopting a soybean pod state joint labeling method based on the three-dimensional point clouds; an improved SoftGroup multi-task learning model is constructed, all the modules work cooperatively according to preset logic, the three-dimensional point cloud of the single plant of soybean serves as input, and target task result output is achieved through point cloud feature extraction and instance clustering; performing multi-task joint training on the improved SoftGroup multi-task learning model, preprocessing the three-dimensional point cloud of the soybean plant to be tested, inputting the preprocessed three-dimensional point cloud into the trained model, performing multi-task reasoning, correcting geometric parameters through PCA fine processing, and counting the pod bearing number; removing pod point cloud from the single-plant soybean three-dimensional point cloud to obtain plant main body point cloud; and extracting a plant skeleton from the plant main body point cloud, and obtaining character data of the plant skeleton.

The invention discloses a field wheat three-dimensional character analysis method based on unmanned aerial vehicle multi-view three-dimensional point cloud, and the method comprises the following steps: S1, shooting a multi-view image and carrying out the three-dimensional reconstruction, and obtaining an original point cloud; performing interpolation segmentation and secondary segmentation, and outputting a plurality of cell point clouds; s2, performing data enhancement, starting model training, and obtaining each 3D bounding box; s3, extracting point cloud data of each planting row from each 3D bounding box by using a row segmentation algorithm, and calculating each static phenotype parameter; and S4, calculating five dynamic phenotypic parameters of the field wheat. According to the invention, based on the 3D wheat plot detection network and the row segmentation algorithm, the point cloud data and the 3D bounding box of the wheat are rapidly extracted, and accurate monitoring and accurate segmentation of the wheat are realized; based on accurate analysis of static phenotypic parameters, a new system of dynamic phenotypic parameters is provided, and based on the dynamic phenotypic parameters, the growth vigor and yield of wheat can be effectively predicted.

The invention discloses a character analysis method based on artificial intelligence. The method comprises the following steps: obtaining user dialogue content; a plurality of artificial intelligence models are constructed, and the plurality of artificial intelligence models comprise a management model and a plurality of scoring models; dividing the character into a plurality of dimensions, and distributing one dimension for each scoring model; inputting the user dialogue content into each scoring model, scoring the user dialogue content by each scoring model according to the distributed dimension, and outputting scoring results of multiple dimensions; and inputting the scoring results of the multiple dimensions into the management model, processing the scoring results of the multiple dimensions through the management model, and outputting a multi-dimensional character scoring report. According to the method, the multiple character dimensions are analyzed through the multiple artificial intelligence scoring models, the scoring report is output through the artificial intelligence management model, and the multi-dimensional character analysis result of the user can be accurately obtained.

The invention discloses an AI psychological counseling method and system for generating an emotional map and a differentiation report, and relates to the technical field of artificial intelligence and psychology crossing. The invention discloses an AI psychological counseling method and system for generating an emotional map and a differentiated report. Through interaction of voice and characters, a visual emotional map is generated, personalized action suggestions are generated, and a visitor report or a psychological counselor report is generated. According to the method, a visual emotional map is generated through voice and character analysis, changes in the emotional map can be seen more visually, meanwhile, a proper psychological AI model is adapted, personalized action suggestions and corresponding related reports are generated, fine emotional fluctuations can be effectively captured through the accurate voice and character capturing capacity, and therefore the emotional effect is improved. A multi-dimensional emotion analysis model is arranged in the system, emotion fluctuation is monitored in real time, a detailed emotion change trend chart is generated, and the scientific basis is provided for psychological counseling.

The application discloses a kind of based on molecular marker clustering analysis's cotton hybrid breeding combination screening method, the screening method includes: step S1: using multi-traits related molecular marker to the gene typing experiment of test cotton variety, data collection and arrangement are carried out;Step S2: excellent allele frequency calculation and feature matrix construction;Step S3: data standardization and principal component analysis dimension reduction;Step S4: optimal cluster number selection and genetic clustering;Step S5: each population character analysis;Step S6: heterosis potential score and combination screening.The technical problem that present cotton breeding parent selection lacks multi-traits comprehensive quantitative evaluation is solved, and the data of breeding decision-making and intelligentization are realized.

The embodiment of the invention provides a photo volume generation method and device based on a large model, and the method comprises the steps: describing the contents of all reconnaissance images in a preset range based on a multi-modal large model according to a first cue word, and generating a detail title of each reconnaissance image; carrying out character analysis on the detail title of each survey image according to a second cue word based on a large language model to obtain a spatial hierarchy corresponding to the detail title; according to the investigation images of different spatial levels, feature extraction is carried out on the investigation images based on a diffusion model network, and duplicate removal is carried out on the same view angle images of the same spatial level; on the basis of a diffusion model network, sequentially segmenting each survey image from the survey image set of the uppermost spatial hierarchy, searching for the most similar survey image in the next spatial hierarchy, and performing spatial matching; based on a result of the spatial matching, a photo volume from a complete spatial hierarchy is generated.

The invention relates to the technical field of character recognition, in particular to a method and device for recognizing characters in a code spraying image in real time, character scene types in the code spraying image are analyzed, character analysis modes matched with the character scene types are selected according to the different character scene types, character recognition is carried out on the code spraying image, and the character recognition efficiency is improved. And for character scenes such as scattered character distribution and short texts, identifying a character identification category of each character in the code spraying image, and for character scenes such as continuous text lines and long texts, identifying a character identification position of each character in the code spraying image, so as to obtain a character identification result matched with the character mode of each character in the code spraying image. The character recognition accuracy and efficiency are improved, the corresponding character recognition results are analyzed based on different character analysis modes to obtain the code spraying quality of the code spraying image, the real-time detection accuracy of the code spraying quality of the code spraying image is improved, code spraying parameters can be adjusted and optimized in a self-adaptive mode, and then the code spraying effect is improved.

The present application relates to the field of image processing, and more particularly to a dish visual processing detection system, comprising: a collection and analysis module, configured to perform quality detection on a randomly selected target detection product and key label image collection, and to detect a collection quality coefficient of each character image in the key label image, and determine the category of the character image according to the collection quality coefficient; a data analysis module, configured to determine a processing mode according to an abnormal character proportion; a secondary processing module, configured to determine a to-be-adjusted character region and perform region adjustment, and determine an image processing mode according to a contrast reference value and a proportion of a reflective region; and a character analysis module, configured to determine a character recognition mode according to an abnormal key coefficient, so that the accuracy of the information recognition result of the key label image with abnormal character images is improved, and the degree of dependence on personnel recognition is reduced.

The invention relates to the technical field of printing character defect detection, in particular to a printing character defect detection device based on a mechanical scanning mechanism, and is used for solving the problems that an existing printing character defect detection mode is low in detection efficiency, high in false detection rate and difficult to meet complex and changeable printing environments and high-quality detection requirements. The device comprises the following modules: a mechanical scanning mechanism, a scanning mechanism control module, a character analysis module, a character defect detection module and a character defect evaluation module, according to the device, the adaptive adjustment of different printed matters is realized by selecting the scanning rate, the accurate acquisition of printed characters is improved, and the defect detection accuracy of the printed characters is improved; according to the device, potential defects can be efficiently and accurately identified by utilizing the defect measurement coefficient and the abnormity early warning coefficient, so that effective detection of various types of printing character defects is realized, and an important basis is provided for quality control of printed matters.

The application relates to a large scene character behavior analysis method, system, device and storage medium, and relates to the field of artificial intelligence. The method comprises the following steps: calibrating scene areas of a plurality of long-range cameras and recording a plurality of area calibration information, wherein the area calibration information comprises coverage areas of the plurality of long-range cameras and object information and object coordinates in the coverage areas; collecting a plurality of scene data by using the plurality of long-range cameras, and synchronizing the plurality of scene data in time; tracking character trajectories in the plurality of scene data by using a pedestrian re-identification model to obtain a plurality of area stay times and a plurality of area images corresponding to the plurality of area stay times; performing multi-modal analysis on the plurality of area images to obtain a plurality of character analysis results; and fusing the plurality of character analysis results to obtain a character analysis comprehensive result. The technical effect of the application is to improve the accuracy of character behavior analysis in a large scene.

The application provides a state machine-based streaming data analysis method, system, device and storage medium. The method comprises: receiving packet streaming data; initializing a state machine and performing character-by-character analysis on the streaming data; when the analysis state of the state machine is a text state, if the analyzed character is a label start symbol, switching the analysis state of the state machine to a label state; when the analysis state of the state machine is the label state, storing the analyzed character in a label buffer until the analyzed character is a label end symbol, ending the character analysis of the current streaming data; querying a predefined configuration mapping table based on the string content of the label buffer to determine a label type; performing a data block operation based on the label type and dynamically appending the recognized non-label character to the current active data block. The application improves the accurate analysis rate of streaming data and the multiplicity and maintainability of the scheme.

A multi-dimensional context and hierarchical authorization dynamic character analysis system and method belong to the field of enterprise-level data authority management, and the method comprises the following steps: a receiving and recognition module dynamically recognizes a time template, a user context template, a function template, a hierarchical authorization template and a parameterized template type through a regular expression; the recursion decoding module avoids infinite recursion and processes any hierarchical nested template; the hierarchical authorization processing module calls a data permission interface to dynamically generate an SQL fragment containing an organization identifier; the parameterization processing module extracts attribute values of an external business object set through a reflection mechanism, and secondary recursive decoding is triggered after the attribute values are assembled into list format character strings. According to the method, dynamic fusion of user, time and organization triple constraints can be realized, single configuration of complex business rules is supported, a new template type can be expanded without hard code modification, a decoding process is traced through a security white list control function call and auditing module, and the flexibility of a permission system and the batch business adaptation capability can be remarkably improved.

The invention relates to the technical field of employment planning, in particular to an AI and character analysis-based occupational planning method and a terminal, which can efficiently screen massive occupational information by issuing character questionnaires, generating character feature vectors and calling an occupational feature library constructed by expert annotations and machine learning. The modal difference between characters and occupational characteristics is eliminated through cross-modal alignment, dynamic attention weights are generated to highlight occupational key character requirements, recommendation is performed according to weighted similarity, the adaptation degree of occupations and user characteristics can be greatly improved, effective occupational recommendation conforming to the user characteristics is provided, short-term demission caused by inconsistent characters is reduced, and the user experience is improved. Not only is job hunting time cost of students reduced, but also recruitment burden of enterprises is reduced.

A computer-implemented method, according to one embodiment, includes receiving, from a user device, a request for determining a final confidence score for an account, and determining a plurality of confidence parameter scores using information associated with the account. A first of the confidence parameter scores is determined based on an analysis of a structure of the account, a second of the confidence parameter scores is determined based on a character analysis of at least the account, and a third of the confidence parameter scores is determined based on an exposure of the account. The method further includes using the determined confidence parameter scores to determine the final confidence score for the account. The determined final confidence score is applied to mitigate fraud.

The invention relates to an abnormal mail detection method and device, computer equipment and a storage medium. The method comprises the following steps: carrying out text anomaly analysis on sending information in a to-be-detected mail by utilizing a preset big language text analysis model, and determining a text anomaly probability of the to-be-detected mail; the big language character analysis model is obtained by performing fine tuning training through an instruction fine tuning data set; the instruction fine tuning data set comprises a mail title, a mail text and a mail abnormal probability; under the condition that the character abnormal probability is greater than a preset probability threshold value, skipping according to an external skipping entry in the to-be-detected mail, and obtaining webpage skipping information corresponding to the external skipping entry; and based on the sending information and the webpage jump information in the to-be-detected mail, performing jump anomaly detection on the external jump entry in the to-be-detected mail to obtain an anomaly detection result of the to-be-detected mail. By adopting the method, mail anomaly detection can be accurately carried out.

The invention provides a streaming data analysis method, system and device based on a state machine and a storage medium. The method comprises the steps of receiving packet streaming data; carrying out initialization processing on the state machine, and carrying out character-by-character analysis on the streaming data; when the analysis state of the state machine is a text state, if the analyzed character is a label start character, switching the analysis state of the state machine into a label state; when the analysis state of the state machine is a label state, storing the analyzed characters into a label buffer area, and ending character analysis of the current streaming data until the analyzed characters are label ending characters; querying a predefined configuration mapping table based on the character string content of the label buffer area to determine a label type; and executing a data block operation based on the label type, and dynamically adding the identified non-label characters to the current active data block. According to the method, the accurate analysis rate of the streaming data and the reusability and maintainability of the scheme are improved.

The application discloses a white oil block stripe improvement method and device based on a jet printer and a medium. The method comprises the following steps: acquiring a production image data template to be jet printed in real time, and acquiring current jet printing times of the jet printer and image processing description information corresponding to the production image data template; performing white oil block and character analysis processing on the production image data template by a preset raster image processor and according to image resolution parameters and pixel sampling point proportion parameters, to obtain a white oil block sampling point graph and a character sampling point graph; acquiring jet printing times corresponding to the current jet printing times of the jet printer; and performing jet printing processing according to the white oil block sampling point graph and the character sampling point graph, to obtain a production image data jet printing image. The method solves the problem that the jet printing of the jet printer cannot improve white oil block stripes, thereby improving the accuracy of the jet printing image, reducing white oil block stripes in the production image data jet printing image, and improving the jet printing quality and accuracy of the production image data jet printing image.

An object of the system according to the embodiment is to investigate the personality of an examinee in detail in an interactive manner and to perform a precise personality evaluation.SOLUTION: A system includes an interaction unit, an analysis unit, an evaluation unit, and a proposal unit. The interaction part interactively investigates the character of the examinee. The analysis unit analyzes the dialogue data collected by the dialogue unit. The evaluation unit evaluates the personality traits based on the analysis result obtained by the analysis unit. The proposal unit proposes the type of occupation or the workplace environment based on the evaluation result obtained by the evaluation unit.SELECTED DRAWING: Figure 1

The invention relates to the technical field of shape acquisition, and particularly discloses planar organ character acquisition equipment which comprises an acquisition box body, an industrial personal computer case is fixedly connected to the upper surface of the acquisition box body, a protective door is mounted on the front surface of the acquisition box body, and a display screen is fixedly mounted on the side surface of the acquisition box body. And a rotating bracket is arranged on the top surface of the cavity of the collection box body. According to the planar organ character acquisition equipment, the acquisition environment and the background cloth state are optimized, the data acquisition quality is guaranteed, the equipment adopts a closed structure to be matched with a constant light supplementing light source, external light interference is avoided, high image definition and consistent light sources are ensured, wrinkles of the background cloth are automatically eliminated through the synergistic effect of the linkage mechanism and the clamping assembly, and the working efficiency is improved. And the arc-shaped positioning clamping blocks and the arc-shaped auxiliary clamping blocks can protect the integrity of the sample and the background cloth, so that the main body of an acquired image is prominent, the background is clean, and high-quality data support is provided for subsequent character analysis.

The invention relates to the field of video stream analysis, in particular to a dynamic video stream real-time character analysis method based on space-time manifold mapping. Comprising the following steps: inputting the preprocessed continuous static image into a deep convolutional neural network, and outputting a continuous feature map; obtaining a spatial-temporal feature vector according to the continuous feature map; inputting the spatio-temporal feature vector as an input point into a spatio-temporal manifold mapping model, and converting the spatio-temporal feature vector into a coordinate matrix in a low-dimensional manifold space; in the low-dimensional manifold space, a dynamic target is identified, and dynamic target information is analyzed; and generating natural language description characters. According to the method, the dimension of the high-dimensional spatial-temporal feature is reduced to the low-dimensional manifold space through spatial-temporal manifold mapping, the feature dimension and the calculation complexity are reduced, the spatial-temporal change relation in the dynamic video can be efficiently extracted in combination with the deep convolutional neural network and the pre-trained large language model, accurate analysis of the object in the dynamic scene is realized, and the accuracy of the dynamic scene is improved. The method has the advantages of efficient real-time analysis, accurate dynamic information extraction and high-quality natural language description.

The invention discloses a mixed language typesetting method and device for embedded equipment, equipment and a medium, and the method comprises the steps: obtaining a to-be-typeset text stream of a mixed language, and dividing the to-be-typeset text stream into a plurality of text fields through a preset Unicode bidirectional algorithm to obtain a text field list; according to the input sequence of the to-be-typeset text stream, character-by-character analysis is conducted on the character fields in the character field list in sequence, so that a font bitmap corresponding to each character is obtained from a preset cache region, according to pre-generated typesetting parameters, a target embedded device is controlled to conduct typesetting rendering on the font bitmap of each character line by line, and typesetting rendering is conducted on the font bitmap of each character line by line. Unified typesetting of the text of the mixed language on the target embedded device is achieved; wherein the typesetting parameters are generated according to a preset alignment rule table; the alignment rule table comprises a baseline offset table and an equal-width proportion table. According to the method and the device, the real-time performance and the visual consistency of the mixed language typesetting of the embedded equipment can be improved.