115 results about "Visual comparison" patented technology

In the field of document processing, format conversion is a common demand, for example, a PDF (Portable Document Format) or a scanning copy is converted into an editable format such as Word, Excel and the like, a traditional method generally adopts an OCR (Optical Character Recognition) technology to directly recognize document content and output a target format, and the method has the defects that the recognition accuracy is insufficient, errors can only be checked and corrected item by item after the target format is converted, and the efficiency is high. In order to solve the problems of lack of visual comparison with an original manuscript, easiness in omission, incapability of adjusting an identification result in real time in a format conversion process and high subsequent editing workload, the invention provides a brand-new document format conversion method, and provides a visual interactive interface by introducing a three-column visual interface and a real-time team correction mechanism. According to the method, the user can check and adjust the recognition result in real time in the conversion process, the accuracy of document conversion is remarkably improved, the flexibility of document conversion is achieved, the user is supported to carry out secondary editing on the content before formal output, visual comparison between the recognized content and an original manuscript is achieved, and the check difficulty is reduced.

The invention relates to the technical field of network communication, and discloses a network communication fault early warning system based on machine learning, which comprises a machine learning module, and the module uses historical operation data to train a model so as to predict the change trend of future communication parameters. The data monitoring module is responsible for monitoring communication parameters of network equipment in real time and performing data cleaning. And the curve fitting module is used for drawing the real-time data and the predicted data into a curve and carrying out visual comparison. And the residual analysis module calculates the residual of the two curves and marks abnormal points. And finally, the early warning notification module judges potential fault risks according to the abnormal points and the evaluation rules and sends early warning information. The system does not depend on a fixed threshold value, can adapt to dynamic changes of a network environment, accurately captures subtle changes and trends of communication parameters, and effectively avoids false alarms and missing alarms. Through the application of the system, the accuracy and timeliness of network communication fault early warning can be remarkably improved, and economic loss and inconvenience are reduced.

The invention discloses a traffic video question and answer method based on inter-frame difference, and the method comprises the steps: innovatively constructing an inter-frame difference extraction module on the basis of an image pre-training model based on the local abrupt change characteristic of a traffic video: effectively capturing the abrupt change characteristic of a moving target through the pixel-level difference calculation of adjacent frames; and designing a gating feature fusion mechanism, dynamically aligning spatial distribution of differential features and visual features on the basis of fusing problem features, and realizing collaborative characterization of static scenes and dynamic changes. In order to enhance cross-modal feature alignment, a hierarchical comparison learning strategy is provided: differential-visual feature comparison is implemented at a bottom layer to enhance local dynamic perception, and text-visual comparison is performed at a high layer to establish global semantic association. Experimental results show that the classification performance is improved on the basis of the SUTD-TrafficQA reference data set.

The invention belongs to the technical field of atmospheric optics, and particularly relates to an atmospheric turbulence phase screen generation method based on complex variable matrix space-time interpolation. According to the invention, a complex variable matrix interpolation algorithm is used to simulate the continuous change of the light field intensity and the random phase screen in the propagation process; comprising the following steps: reflecting time continuity of a turbulent phase screen by using a random complex variable matrix according to a frozen flow hypothesis; based on a power spectrum inversion method, an atmospheric turbulence phase screen is generated; performing light field diffraction transmission simulation by using a Fresnel method; and displaying the change of results in different time steps by utilizing visual comparison. According to the invention, the transmission performance of the optical signal in a changing environment can be effectively improved, and the calculation complexity is greatly reduced.

Upon an attempt to access a service of a third-party server, full-duplex password-less authentication provides a one-time password to the user displayed at the client device and at a mobile device associated with the user. The user verifies the access by comparing the one-time password displayed at the mobile device and the one-time password displayed at the client device. In some embodiments, the one-time password is displayed as a picture while in other embodiments, a combination of a picture the one-time password is displayed as a picture and a set of alphanumeric characters for ease in making the comparison. The user determines whether to accept or deny the authentication sequence after a simple visual comparison.

Magnetic core detection equipment capable of achieving multidirectional visual comparison and identification comprises a magnetic core straightening conveying assembly and a rotary disc conveying mechanism, a plurality of magnetic core positioning grooves are evenly distributed in the rotary disc conveying mechanism around the center circumference of the rotary disc conveying mechanism, and control structures used for clamping or adjusting magnetic cores are arranged in the magnetic core positioning grooves. A plurality of photographing mechanisms used for photographing different faces of the magnetic cores are arranged on the rotary disc conveying mechanism, an angle control mechanism used for driving the control structure to drive the magnetic cores to conduct angle rotation adjustment is arranged on the rotary disc conveying mechanism, and a transverse channel is arranged above one magnetic core positioning groove; in addition, the control structure can control clamping, loosening and angle rotation of the magnetic core, clamping conveying, loosening blanking, clamping rotation and other actions of the magnetic core in the detection process can be completed conveniently, efficient magnetic core detection work is achieved, input of manpower and material resources is reduced, magnetic core defects are compared and analyzed in cooperation with vision, automation is better, and efficiency is higher.

The invention discloses a graph neural network interpretability method based on key sub-graph retrieval, and provides a brand-new global and non-model-dependent key sub-graph retrieval method for serving decision interpretation aiming at the problem that an existing graph neural network (GNNs) interpretability strategy often neglects the importance of sub-graphs. According to the strategy, based on node representation generated by GNNs, a core sub-graph influencing graph label attributes can be accurately identified. And through prediction experiments on a BA3 data set and a Mutageicity data set and visual comparison evaluation with other mainstream interpretation strategies, the superior accuracy and excellent generalization of the method provided by the invention are verified.

The invention relates to an MIDI cloud music system with intelligent analysis and teaching functions and a teaching method, and relates to the technical field of musical instruments. The system comprises a user terminal module which obtains playing data and uploads the playing data to a cloud end; the cloud server module is used for analyzing the MIDI data and generating feedback information and accompaniment data; the intelligent analysis and feedback module performs analysis according to the playing pitch, rhythm and strength, and generates error correction prompts, playing suggestions and visual comparison data; the automatic arrangement and accompaniment module generates an accompaniment track and an arrangement structure according to the selected music style, rhythm and playing characteristics, and sends the track and the structure to the user terminal module for playing; the personalized teaching management module is used for dynamically adjusting teaching contents and difficulty levels and pushing teaching resources according to the learning progress and the playing level; and the database module is used for storing playing records, intelligent analysis results, automatic arrangement information and personalized teaching resources. The application has the effect of improving the application value of the cloud music system and the user experience.

The invention discloses a medical aid decision-making system based on a multi-modal large model, and relates to the technical field of medical artificial intelligence, and the system comprises a multi-modal data fusion module, a treatment scheme coding and management module, a prognosis prediction model module, a scheme simulation and deduction module, and a visual comparison module. The comprehensive state vector of a patient and the action vector of a candidate treatment scheme are jointly calculated, multiple long-term prognosis indexes after the scheme is executed are directly simulated, the model converts the treatment scheme into a computable variable by learning rules in historical treatment data, probabilistic prediction is carried out on a future result, and the prediction accuracy is improved. This enables a doctor to clearly see the risk brought by different selections and the long-term influence of treatment changes before making a decision, thereby converting the decision mode from experience-based inference to future simulation-based anticipation.

The invention relates to the technical field of visual position recognition, and particularly discloses a multi-modal visual position recognition reordering method and system based on guidance, and the method comprises the steps: obtaining a query image, and retrieving a plurality of candidate images based on a pre-trained visual basic model and the query image; constructing a composite multi-modal prompt object, wherein the composite multi-modal prompt object comprises an image pair formed by the query image and the current candidate image, and an instruction text used for guiding a multi-modal large language model to perform visual comparison; outputting a structured similarity judgment result, wherein the result comprises a quantitative similarity score; and sorting based on the similarity scores corresponding to all the candidate images, and determining the candidate image with the highest score as an optimal matching result. Through combination of guiding type prompt engineering and structured output, an intermediate text generation link is avoided fundamentally, and the calculation efficiency is improved while the fidelity of all original visual information is reserved.

The invention provides a document information intelligent extraction and comparison method and system, and relates to the technical field of document intelligent processing analysis, and the method comprises the following steps: obtaining a comparison document and a to-be-compared document, and converting the comparison document and the to-be-compared document into PNG images; selecting a comparison area and a to-be-compared area, and preprocessing the PNG image to obtain a comparison area image and a to-be-compared area image; performing table recognition based on the comparison region image and the to-be-compared region image; according to the table recognition result, extracting a comparison text and a to-be-compared text; semantic understanding is carried out based on the comparison text and the to-be-compared text; and performing retrieval and text difference comparison according to the semantic understanding result to obtain row-level difference and character-level difference, and generating a visual comparison report based on the row-level difference and the character-level difference. The method can adapt to processing of multiple types of documents, is high in precision of complex table recognition and semantic understanding, supports analysis and comparison of specified areas or contents, and is visual in comparison result difference display.

In example embodiments, techniques are provided for visually comparing digital content for an infrastructure project according to time using 4-D construction modeling software. The 4-D construction modeling software includes a cloud-based 4-D comparison service and a local 4-D modeling client. The 4-D comparison service includes a digital content alignment service and a 4-D difference engine. The digital content alignment service aligns different pieces of digital content and produces views that provide visual comparison between different pieces of digital content. The 4-D difference engine automatically determines differences between different pieces of digital content. The 4-D modeling client includes a 4-D comparison user interface (UI) process that receives user input used to generate, and then displays a generated visual comparison between different pieces of digital content. The 4-D comparison UI utilizes time control channels for selecting digital content and comparison controls for selecting a type of visual comparison.

The present invention discloses an FPC board detection device provided with a visual detection device, comprising a conveying mechanism, wherein a side positioning structure for positioning an FPC circuit board is arranged on the conveying mechanism, a lifting component is arranged above the conveying mechanism, a photographing mechanism for detecting the appearance of the FPC circuit board is arranged on the lifting component, and two retractable positioning mechanisms for positioning the front and rear ends of the FPC circuit board are arranged on the conveying mechanism below the photographing mechanism, which automatically fine-tune the front, rear, left and right positions of the FPC circuit board to make it centered in the middle of the photographing position. At the same time, when the front and rear positioning mechanisms are positioning, the photographing device follows and descends to the upper surface of the FPC board for clearer photography. After visual comparison between multiple groups of photographed pictures and terminal qualified pictures, it is quickly detected whether the FPC board is qualified, the detection efficiency is high, and the labor force is reduced.

The invention provides a real-time interaction-oriented index aperture unified processing method, computer equipment and a computer program product, and belongs to the technical field of artificial intelligence, and the method comprises the following steps: receiving and analyzing a dictation request of a user, and generating a structured context state vector; based on the context state vector, obtaining a plurality of candidate denominator calibers matched with the dictation request and a basis thereof from a calibers knowledge base; generating a constraint strategy set based on the context state vector, the candidate denominator caliber and the basis thereof; based on a constraint strategy set, performing consistency judgment through the candidate denominator calibers to determine a unique denominator caliber; based on the unique denominator caliber, generating an index description format final draft, and embedding a restriction rule for prohibiting visual comparison of inconsistent indexes into the index description format final draft; and rendering in a data visualization interface according to the index description format finalization and the restriction rule thereof, and intercepting and guiding the parallel display operation violating the restriction rule in real time.

This application provides a method, computer device, and computer program product for unified processing of indicator definitions for real-time interaction, belonging to the field of artificial intelligence technology. The method includes: receiving and parsing a user's verbal request and generating a structured contextual state vector; based on the contextual state vector, obtaining multiple candidate denominator definitions and their basis from a definition knowledge base that match the verbal request; generating a set of constraint strategies based on the contextual state vector, the candidate denominator definitions, and their basis; based on the set of constraint strategies, performing consistency adjudication through the candidate denominator definitions to determine a unique denominator definition; generating a finalized indicator description format based on the unique denominator definition, and embedding a restriction rule for prohibiting the visual comparison of inconsistent indicators into the finalized indicator description format; rendering the finalized indicator description format and its restriction rule on a data visualization interface, and intercepting and guiding parallel display operations that violate the restriction rule in real time.

The invention discloses a crack image segmentation method based on a bidirectional structure sensing cascade network. The crack image segmentation method comprises the steps of performing preprocessing based on a collected crack image; performing block processing and position coding on a crack image in the preprocessed data to obtain coded image data; obtaining a BiScaNet network model based on a mixed objective function combining binary cross entropy loss and Dice loss; after feature extraction is carried out on the coded image data, fusion and decoding are carried out on features of different scales, and a pixel-level segmentation image is generated; and carrying out an ablation experiment and outputting evaluation indexes, and carrying out visual comparison on the pixel-level segmentation image and segmentation images of other models. According to the invention, the scanning strategy in the Mama architecture is improved, and the BiSS module is provided, so that the modeling capability of crack multi-direction extension and complex topological structures is remarkably enhanced.

The utility model relates to a flowmeter display platform, which belongs to the technical field of flowmeter display and comprises a display rack, a display groove is arranged in the display rack, a water pipeline is fixedly mounted in the display groove, different flowmeter bodies are fixedly mounted at different positions of the water pipeline, and the number of the flowmeter bodies is nine. Wherein the four flow meter bodies are connected in parallel, a water valve is installed on one side of each flow meter body connected in parallel, a water tank a is embedded in the display rack and located on the left side of the display groove, and a water tank b is embedded in the display rack and located on the right side of the display groove. The beneficial effects of the utility model lie in that the flow can be controlled by adjusting the water valve, and then the reading change of the flow meter body can be observed, and in the exhibition item, various flow meter bodies are displayed together to form mutual reference and comparison, so that audiences can understand the working principles of different flow meter bodies in a visual comparison mode, and the working efficiency is improved. And an interaction mode of adjusting the flow of the water pipeline is integrated to attract audiences.

The invention relates to the technical field of front-end development, and discloses a page exception judgment method and device, equipment, a medium and a program product, document object model nodes and document object model elements of a target page are obtained, and a target node is screened out from the document object model nodes according to business data corresponding to the target page; performing structural feature analysis on the target nodes according to a preset measurement index to obtain a structural feature analysis result, and performing semantic analysis on the document object model element in each target node to obtain a semantic analysis result; and determining whether the target page is abnormal or not based on the structural feature analysis result and / or the semantic analysis result. Structural feature analysis is performed on the target node so as to capture the structural distortion problem caused by compatibility, dynamic loading or rendering errors, and the limitation that visual comparison lags behind and is easily influenced by UI iteration is broken through; and semantic analysis is performed on the DOM elements in the nodes, so that the defects of narrow attribute monitoring range and low accuracy are overcome.

Systems, methods, and computer program products for identifying a candidate product in an electronic marketplace based on a visual comparison between candidate product image visual text content and input query image visual text content. Unlike conventional optical character recognition (OCR) based systems, embodiments automatically localize and isolate portions of a candidate product image and an input query image that each contain visual text content, and calculate a visual similarity measure between the respective portions. A trained neural network may be re-trained to more effectively find visual text content by using the localized and isolated visual text content portions as additional ground truths. The visual similarity measure serves as a visual search result score for the candidate product. Any number of images of any number of candidate products may be compared to an input query image to enable text-in-image based product searching without resorting to conventional OCR techniques.

The invention relates to a portable automobile window film haze testing device which comprises a shell, a lampshade and glass, a groove and a hole are formed in the shell, a support is further arranged on the lower portion of the shell and fixedly connected with the shell, and a cabin is arranged in the shell; the lampshade is provided with a socket hole, a button and an LED lamp, the socket hole, the button and the LED lamp are electrically connected, and extinction velvet is arranged in the cabin body. The test cost can be reduced, the customer experience can be improved, real vehicle use environment simulation is adopted, the test result is closer to actual use, the vehicle owner can understand the test result more easily, the persuasion of the window film performance is enhanced through visual comparison, and the situation that the definition is affected and then the safety is affected due to the haze effect of the vehicle window film is reduced.

The invention relates to the technical field of intelligent manufacturing, in particular to an intelligent manufacturing unit-oriented integrated industrial robot control system, which comprises an integrated control center, an initial interaction module, a task scheduling module, a dynamic collaboration module, a collaboration presentation module and an early warning response module, according to the method, identification, matching and testing of robot communication protocols are automatically completed, the equipment access time is remarkably shortened, meanwhile, it is ensured that all robots can stably complete data interaction, system access dead angles caused by protocol incompatibility are avoided, candidate equipment is screened based on production task requirements and the real-time states of the robots, and the production efficiency is improved. Meanwhile, the task execution error probability is reduced through multi-dimensional comprehensive evaluation and in combination with a load capacity deep screening robot, in addition, in the material grabbing link, center deviation is judged through visual comparison, re-grabbing or abnormal early warning is automatically triggered, production stagnation caused by material grabbing errors is reduced, along with reasonable path planning, the production efficiency is improved, and the production cost is reduced. And the robot driving path safety is ensured.

The present invention claims protection for a scheduling and management system for material distribution services based on the collaboration of unmanned aerial vehicles and vehicles, which relates to the technical field of material distribution. It includes: a material management module, an algorithm operation module, an algorithm performance comparison module, an algorithm comparison visualization module, and a distribution management module. The material management module is used to manage materials; the algorithm operation module is used to process and analyze the selected logistics data and obtain the transportation route information of the collaboration between unmanned aerial vehicles and vehicles in combination with the actual traffic topology, and display the route map under different numbers of users; the algorithm performance comparison module is used to compare the performance of other algorithms; the algorithm comparison visualization module is used to display and draw a visual comparison map of the routes for comparison in different traffic environments and at different collection points; the distribution management module is used to select the distribution materials, and obtain the distribution site and distribution logistics information in real time, and visually display them in the form of charts.

The present invention discloses a non-contact pen-holding posture calibration method and system based on horizontal and vertical stroke decomposition, which belongs to the field of wireless perception and human-computer interaction. The present invention obtains channel state information through a universal WiFi network card device, realizes stroke segmentation and amplitude data analysis through time-frequency analysis and mutation point detection; utilizes adversarial learning to extract position robust features, and performs pen-holding posture recognition through a bidirectional long short-term memory network model enhanced by an attention mechanism; adopts horizontal and vertical probability weighted calculation to calculate the similarity with the standard posture library, and when the similarity is lower than the threshold, triggers a sound warning and generates a visual comparison calibration suggestion. The present invention can overcome the influence of different handwriting positions and differentiated handwriting contents, realizes accurate recognition and real-time feedback of non-contact pen-holding posture, and has the advantages of strong position robustness and universal applicability.

The invention discloses a visual comparison method of a graphic model file and a computer readable storage medium, and the visual comparison method of the graphic model file comprises the steps: loading a graphic model file needing to be compared, and carrying out the structural reduction of the graphic model file to obtain model data; comparing the model data, searching different model objects in the graphic model file to be compared, and storing the different model objects in a different model object list; and respectively carrying out visual rendering on model data of the graphic model files needing to be compared in the graphic view, and differentially displaying the difference model objects selected in the difference model object list in the graphic view. Compared with a traditional comparison method based on text content, the method provided by the invention provides a more intuitive visual comparison means for the graphic model file, and the method not only improves the modeling efficiency, but also remarkably reduces the time required for troubleshooting and solving problems.

A system and method for selecting image sequences of an echocardiographic study for visual comparison by a user. The method includes receiving image sequences from a current study and a previous study to be compared. The method includes determining relevant views of the image sequences in the current study and the previous study, and selecting image sequences of relevant views in the one of the current study and the previous study for comparison. The method includes displaying image sequences from one of the current study and the previous study to a user via an interactive comparison panel. The method includes: determining which image sequence in relevant views of the other of the current study and the previous study most closely matches the selected image sequence; and displaying the image sequence from the other of the current study and the previous study that most closely matches the selected image sequence via the interactive comparison panel.

The utility model relates to the technical field of detecting the installation direction of a terminal by utilizing optics, in particular to detection equipment for judging the installation direction of the terminal by utilizing mechanical vision. The image acquisition component is arranged on one side of the moving component and can shoot and acquire a to-be-verified image, and the shooting direction is perpendicular to the mounting direction of the female terminal; a plurality of detection parts which correspond to the pins with different installation directions and are slidably arranged on the moving part along the installation direction of the female terminal; the power component is used for driving the moving component to move; the image processing terminal stores a reference image for mounting the female terminal and is used for receiving the to-be-verified image and comparing the reference image with the to-be-verified image; according to the utility model, a clear image to be verified can be obtained, the installation orientation of the internal pin of the female terminal can be rapidly judged, and the visual comparison efficiency of the installation orientation of the internal pin of the female terminal is improved.

1. Name of the Design Product: Visual Comparison Interactive Graphical User Interface for Version Differences of Electronic Devices. 2. Use of the Design Product: An electronic device. 3. Design Key Points of the Design Product: Lies in the graphical user interface. 4. Picture or Photo that Best Illustrates the Design Key Points: The front view of Design 1. 5. Design 1 is designated as the basic design. 6. Use of the Graphical User Interface: The interface is used to display version comparison result information, quickly locate difference point information, etc. 7. Human-Computer Interaction Method of the Graphical User Interface: In the front view of Design 1, the information bar at the top of the interface correspondingly displays several modification styles such as "Delete, Add, Modify", and on both the left and right sides below it are comparison modules for different version information. When the user double-clicks on any node in the comparison module, the interface jumps to the interface change status diagram of Design 1. In the interface change status diagram of Design 1, the comparison difference points of the version within the node are clearly shown. The human-computer interaction method of Design 2 is the same as that of Design 1.

The invention relates to the technical field of computer-aided engineering, and provides a BREP-based CAD model comparison method, which comprises the following steps of: carrying out visual comparison between two target CAD models; if the visual comparison results of the two target CAD models are inconsistent, determining that the two target CAD models are different models; if the visual comparison results of the two target CAD models are consistent, comparing boundary representation BREP data of the two target CAD models; and if the boundaries of the two target CAD models indicate that the BREP data comparison results are consistent, determining that the two target CAD models are the same models. According to the technical scheme, the CAD models can be objectively and accurately compared.

A whole building or campus entity, energy management and energy control optimization, data presentation and visualization dashboard system including a computer processor service provider system, wherein the computer processor service provider system comprises a cloud-based computer processor service provider system comprising: at least one application server comprising the whole building or campus entity, energy management and energy control optimization system executing in the cloud-based computer processor service provider system; and an onsite energy controller coupled to the cloud-based computer processor service provider system by an electronic communications network; wherein said onsite energy controller comprises: at least one electronic computer processor; and at least one electronic computer memory coupled to said at least one electronic computer processor; wherein said at least one electronic computer processor of said onsite energy controller is configured to: reduce energy and demand costs through whole building or campus entity energy management of a building or campus entity load including by being configured to at least one or more of: optimize energy usage; optimize energy generation; maximize demand reduction and utility savings through optimal control of: the energy usage, and the energy generation, or reduce energy costs; and use of at least one energy storage system device; simulate said building or campus entity load by use of at least one or more of: predictive analytics; or receive user inputs of an event schedule that impacts said building or campus entity load; perform optimization at the start of the billing cycle to establish a demand reduction target achievable with the at least one energy storage system device implemented; set a periodic schedule for discharge of the at least one energy storage system device and allocate a portion of energy storage capacity to a reserve to be used in the event of volatility in said building or campus entity load; determine whether there is unused capacity of the at least one energy storage system device at an end of a period, and responsive to the determination, redistribute any of the unused capacity to a remaining period of the periodic schedule; and reset the demand reduction target for each time-of-use period to what was achieved in the previous period for the same billing period; and wherein said at least one electronic computer processor of said onsite energy controller is configured to: display output interactively via an electronic dashboard graphical user interface (GUI) comprising at least one GUI element enabling interactive user receipt of input and provision of output relating to at least one or more of the following: a heat map; a financial energy savings calculation; a color coding of energy production depicting at least one or more of: actual energy production, weather normalized energy production, or predicted energy production; or a color coding of energy usage performance based upon an automatically calculated comparison of expected energy production to actual energy production; a color coded representation of monetization comprising at least one or more of: multi-colored lines on a graphically displayed chart metaphor GUI element; or multiple revenue streams; visual comparison of data between different time periods; or artificial intelligence (AI)-enabled data showing at least one or more of: anticipated production data or anticipated consumption data.

The invention discloses a monitoring carrier and a method for monitoring laser energy stability, the monitoring carrier comprising: a core layer; a first copper foil disposed on the core layer; and a second copper foil disposed at the bottom of the core layer, wherein at least one characteristic structure for simulating a non-uniform material structure of the printed circuit board is pre-embedded in the core layer. According to the monitoring carrier and the method for monitoring the laser energy stability, by establishing a benchmark standard atlas database based on destructive physical analysis and implementing periodic visual comparison diagnosis, tiny attenuation of the laser process capability in the aspects of material interface treatment, depth control, heterogeneous material processing and the like can be directly and effectively found.