99 results about "Target type" patented technology

This application relates to a method, apparatus, computer device, and storage medium for displaying medical images. In response to an instruction to acquire a first image sequence, a target type label for the first image sequence is determined. From a sequence of images to be identified containing various type labels, at least one second image sequence associated with the target type label is acquired. The first image sequence and at least one second image sequence are then displayed. Each type label is obtained based on a classification result determined according to each image sequence to be identified and a preset classification model. The embodiments of this application employ a preset classification model to automatically identify the image sequences to be identified, obtaining the classification results of each image sequence to be identified, thereby obtaining the type label for the image sequences to be identified. Compared to relying on manually entering tag information for medical images, the method proposed in this application can significantly reduce the error probability of acquiring the second image sequence based on the target type label, thereby greatly improving the accuracy of image sequence acquisition.

Embodiments of the present application relate to a document-based operation method and device, electronic equipment and storage medium, the method comprising: in the case of detecting a first trigger operation on a target type hyperlink in a document, determining a function corresponding to the hyperlink; and performing an operation corresponding to the function. Thus, a trigger entry in the form of a hyperlink can be provided to a user in a document to trigger a specific function, user operation is more convenient, and user experience is better.

This invention belongs to the technical field of natural language processing and dialogue systems. It discloses a multi-target dialogue recommendation method based on hierarchical prompt optimization, including hierarchical target label modeling, learnable prompt embedding initialization, graph attention enhancement mechanism, multi-task learning framework, and response generation model. Hierarchical target label modeling encodes the hierarchical relationships between target labels into a graph structure. Learnable prompt embedding initialization integrates label semantics and deep hierarchical information into the encoder input space. The graph attention enhancement mechanism enhances the label vector structure perception ability through information transmission and aggregation between label prompts. The multi-task learning framework jointly optimizes masked language modeling and hierarchical multi-label classification. The response generation model uses the predicted target as a control signal to guide generation. This method effectively models the hierarchical dependency relationship between target types and target entities, enhances the structure perception ability of label vectors, maintains the model's language understanding ability, and improves the accuracy of target prediction.

This invention relates to a method and system for tracking partially distinguishable group targets based on augmented label RFS. The method includes: acquiring a set of measurement points; judging the set of measurement points to obtain measurement points for distinguishable and indistinguishable targets; clustering the measurement points of indistinguishable targets and calculating the centroid as equivalent measurement; recursively estimating the target state using a GALMB filter to obtain a set of estimated target states; updating the target type in the augmented label using the Bernoulli elements obtained from updating the equivalent measurement in the GALMB filter update step; dividing the estimated target state set into group structures and updating the group structure part in the filter label to obtain target trajectory, state, and group structure information. This invention is applicable to complex application scenarios such as air defense early warning, air traffic control, and battlefield situational awareness.

The application discloses a method and device for selecting automobile plate, an electronic device and a storage medium, wherein the method comprises: obtaining target type information and a target performance range of a target automobile plate to be selected; obtaining a plurality of candidate automobile plates conforming to the target type information, and obtaining mechanical index parameters corresponding to each candidate automobile plate; for each candidate automobile plate, predicting a predicted performance range of the candidate automobile plate by using a trained performance prediction model based on the target type information and the mechanical index parameters; and selecting the target automobile plate from the plurality of candidate automobile plates based on the target performance range and the predicted performance range corresponding to each candidate automobile plate. The accuracy of selecting the automobile plate can be improved.

The application discloses a large language model backdoor attack detection method, and belongs to the field of artificial intelligence security. In view of the problem that existing defense methods are difficult to be applied to generative large language models and have large calculation overhead, the application is based on the first discovered "sequence locking" phenomenon, that is, when generating an attack target, the backdoor model will output an abnormally high and consistent word element confidence, forming a deterministic path without branches. The method detects high-confidence sequences that continuously exceed the length threshold L and have a probability higher than the threshold P by monitoring the Top 1 probability of the model output sequence in real time, thereby determining the backdoor attack. The method only needs to access the Top 1 probability of the model in a black box, does not need complete Logits vectors or auxiliary models, almost does not introduce additional delay and performance overhead, is suitable for real-time interaction and large-scale deployment scenarios, and can efficiently detect various backdoor triggering modes and target types.

This application provides a remote sensing data processing method and system, electronic device, and storage medium, relating to the technical field of remote sensing data processing. The remote sensing data processing method includes: acquiring target remote sensing data from a target satellite constellation based on received user data requests; parsing the sensor type corresponding to the target remote sensing data to determine at least one target type of remote sensing data; determining the metadata information of the target type remote sensing data; performing block processing on the target type remote sensing data based on the metadata information to obtain multiple sub-target type remote sensing data; and performing distributed parallel computation on the multiple sub-target type remote sensing data to obtain the target output result. This application enables online access and acquisition of remote sensing data by parsing user data requests online, avoiding local downloads and offline processing. By migrating remote sensing data and algorithms to the cloud, it solves problems such as long processing time, low processing efficiency, and high local resource consumption costs.

The application discloses a kind of detection method, device, electronic equipment and readable storage medium of differential mechanism in vehicle.The method comprises: obtaining the operating state signal of disconnecting component of differential mechanism in vehicle;Based on operating state signal, determine the target response duration of disconnecting component executes target type operation, wherein, target type operation is one of connecting operation and disconnecting operation, connecting operation is used to control motor and power system to establish power connection, disconnecting operation is used to control motor and power system to disconnect power connection;Based on target response duration and the reference response duration corresponding to target type operation, determine the working state of differential mechanism, wherein, working state is used to indicate whether differential mechanism exists fault;In response to working state indicating that differential mechanism exists fault, generate early warning information, wherein, early warning information is used to at least indicate to maintain differential mechanism.The application solves the technical problem that differential mechanism in vehicle cannot be detected in time.

This invention relates to the field of remote sensing image technology and discloses a method and system for fine-grained target detection in remote sensing images. The method includes the following steps: S1, obtaining task requirements; S2, acquiring remote sensing images; S3, querying a fine-grained target template library based on target labels to determine whether there is an instance template corresponding to the target type. If no instance template exists, the required target name is used directly, and a basic dynamic prompt is generated in conjunction with a remote sensing knowledge base. Zero-sample target detection is then performed on all slices. Otherwise, proceed to S4; S4, template instance feature parsing; S5, dynamic prompt instantiation generation; S6, coupled matching and detection; S7, extracting the corresponding slice information, generating and outputting geographic information products and detection reports conforming to a predetermined format.

The application discloses an information collection method, which comprises the following steps: receiving target configuration information sent by a first base station; wherein the target configuration information at least comprises a target type of channel information to be collected; receiving a reference signal sent by a second base station; collecting target channel information corresponding to the target type based on the target configuration information and the reference signal; and sending the target channel information to a third base station. The application also discloses a terminal and a computer readable storage medium.

This application provides a method, device, medium, and program product for determining the envelope region of reference points for parts. The method includes: obtaining the initial spatial coordinates of all reference points for a target type of parts; determining the first reference coordinates corresponding to all parts and the second reference coordinates corresponding to each part; then determining the offset of the initial spatial coordinates based on the first and second reference coordinates, and adjusting the initial spatial coordinates using the offset to obtain the corrected spatial coordinates of each part reference point; finally, combining the reference point types of all parts and the corrected spatial coordinates to determine the envelope region of all reference points for the target type of parts. This application's embodiment adjusts the initial spatial coordinates based on the offset of each part and calculates the envelope region of the part reference points based on the corrected spatial coordinates, achieving a part envelope region that approaches global optimum, significantly reducing the structural redundancy and development cost of part measurement supports.

The application discloses a network point reservation method and device, and receives a network point reservation request sent by a terminal device of a user through a 5G message. The network point reservation request is analyzed to obtain key information such as location information of a current location of the user, a target type service, a target queuing mode and a target sorting mode. The user can determine the target type service, the target queuing mode and the target sorting mode according to self demands and include the information in the network point reservation request. After the key information is obtained, a first preselected network point for handling the target type service within a preset range around the current location of the user is determined. Then, a second preselected network point meeting a preset sorting range is determined from the first preselected network point according to the target queuing mode and the target sorting mode. Finally, a target network point is determined from the second preselected network point, a queuing request is sent to the target network point, and the reservation of the target network point is completed. In this way, the selected target network point can meet the expectation of the user, and the experience of the user is improved.

The application discloses a text recognition method, device, equipment, medium and product. The text recognition method comprises the following steps: setting a target recognition template for a target type of certificate; obtaining a target certificate image corresponding to the target certificate, performing text recognition on the target certificate image, and obtaining an initial recognition result; determining a plurality of pairs of target text regions and target anchor point regions matched with text information; determining mapping region position information corresponding to a mapping region of a to-be-recognized region in the target certificate image according to target region position information; determining a first text region intersecting with the mapping region from a plurality of text regions; obtaining first text information corresponding to the first text region from the initial recognition result, and outputting a recognition result corresponding to the target certificate image according to the first text information. According to the embodiment of the application, the text recognition of different types of certificates can be performed based on a general recognition model, and a large amount of time and resources are saved.

The application relates to the technical field of electric power safety monitoring, and discloses an electric power scene intelligent monitoring and picture intelligent screening and saving system and method, which aims to solve the problems of low recognition accuracy, non-targeted picture saving, incomplete feature extraction, single retrieval function and poor tool collaboration in the existing electric power monitoring technology. The application is equipped with a YOLOv7 target recognition model optimized according to an electric power scene through a C++ and Python collaborative architecture, high-precision target detection and multi-dimensional feature extraction of video frames are realized, meanwhile, an intelligent picture screening logic based on a target type and a time interval is designed, a multi-dimensional data retrieval function is matched, and a full-process automatic monitoring system of video decoding-recognition extraction-screening saving-data retrieval is formed. Experiment verification shows that the target recognition accuracy of the application can reach 98.5%, the invalid picture redundancy rate is greatly reduced, and the intelligentialization and efficiency of electric power line and substation safety early warning are improved.

A few-shot, unsupervised image-to-image translation (“FUNIT”) algorithm is disclosed that accepts as input images of previously-unseen target classes. These target classes are specified at inference time by only a few images, such as a single image or a pair of images, of an object of the target type. A FUNIT network can be trained using a data set containing images of many different object classes, in order to translate images from one class to another class by leveraging few input images of the target class. By learning to extract appearance patterns from the few input images for the translation task, the network learns a generalizable appearance pattern extractor that can be applied to images of unseen classes at translation time for a few-shot image-to-image translation task.

The application discloses a kind of multi-modal named entity recognition methods for multi-graph scene, belong to computer vision and multi-modal information extraction technical field, including: constructing entity cross-graph candidate visual region set and determining initial visual support weight;Based on entity type label, entity condition multi-instance supervision is carried out, and false matching is inhibited by non-entity negative sample constraint;Based on the entity type prediction result of set, visual region disturbance is carried out, and contribution degree and discriminativeness are determined according to target type prediction score change;Based on contribution degree and discriminativeness, visual reconstruction is guided by contribution, and support maintenance ability is determined according to target type prediction score retention degree;Further, visual contribution calibration score is generated to calibrate aggregation, and calibrated cross-graph entity level visual representation is directed to text representation, to obtain multi-modal named entity recognition result.The method can improve the reliability and accuracy of entity recognition in multi-graph scene.

This application relates to a method, apparatus, device, medium, and program product for super-redundancy processing. The super-redundancy processing method includes: identifying super-redundant data blocks in a storage device; determining the target metadata instance corresponding to the super-redundant data block; determining the service status of the target metadata instance in the corresponding target storage pool within the storage device based on the target type to which the target metadata instance belongs; and controlling the target storage pool to be in a service state. This method can maintain the availability and business continuity of the storage system to a certain extent, and improves the flexibility of the super-redundancy processing mechanism.

The application discloses a low-code platform variable type automatic inference method and device. The method comprises the following steps: screening a variable with an automatic type identifier in a low-code platform to obtain a target variable; obtaining a default value inference result based on a type inference rule; traversing a structured model corresponding to a business logic and declaration information to obtain a node assignment inference result; obtaining a dependency relationship inference result through a multi-round iteration strategy recursive derivation; integrating the default value inference result, the node assignment inference result and the dependency relationship inference result to determine an initial type of the target variable; storing the initial type based on inference trace information and realizing change notification to finally obtain a target type. Through multi-stage inference, expression type calculation, chain dependency processing and other designs, the application does not need to manually specify the type, supports complex scene inference, maintains type safety, reduces use and maintenance costs, and improves the development efficiency and ease of use of the low-code platform.

A method for shot pattern selection when combating targets with an impact system includes estimating the position of the target, estimating the speed of the target, categorizing the type of target, selecting the type of shot pattern, and firing projectiles based on the selected shot pattern. A fire-control system and an impact system are also provided.

This invention discloses a radar target classification method and apparatus based on time-frequency feature fusion. In this method, radar echo signals are acquired and preprocessed to obtain multiple frames of target signals. Based on pre-defined micro-Doppler physical models for different target types, time-frequency analysis is performed on the multiple frames of target signals to extract multi-dimensional feature vectors containing micro-Doppler features. Target types include pedestrians, motorcycles, cars, and buses. Variance analysis is used to select features from the multi-dimensional feature vectors, retaining a pre-defined number of key features with the strongest discriminative power to form optimized feature vectors. The optimized feature vectors are then input into a pre-trained support vector machine classification model to obtain the target classification result. This invention achieves accurate and rapid identification of millimeter-wave radar targets through precise micro-Doppler physical modeling and multi-dimensional feature fusion, improving the accuracy and robustness of target classification in complex environments.

This application discloses a data processing method, apparatus, and storage medium. It involves acquiring transaction information; then treating transaction objects and transaction platforms as network nodes, and connecting these nodes based on the transaction information to obtain a heterogeneous network; performing node traversal within the heterogeneous network to obtain meta-paths; and representing these meta-paths using vectors to obtain node vectors. Finally, it classifies transaction objects based on the distance between node vectors to obtain objects of the target type. This achieves an AI-based merchant identification process. Due to the wide distribution of transaction platforms, the coverage of transaction objects is guaranteed, and the vector representation involved in the identification process is based on the occurrence of transaction behavior, improving the accuracy of vector representation and object type identification.

Embodiments of the present application provide a fraud identification model construction method, a fraud identification method and equipment. The method can include: obtaining a plurality of fraud case description texts; processing the plurality of fraud case description texts using a large model to obtain common feature description information corresponding to fraud cases of different types; retrieving from a modeling expert knowledge base according to the common feature description information of a target type of fraud case to obtain a modeling scheme corresponding to the target type of fraud case, wherein the modeling expert knowledge base stores a corresponding relationship between case features and modeling schemes, and the target type is any of the case types contained in the plurality of fraud cases; and constructing an identification model for fraud of the target type according to the modeling scheme. In this way, a precise identification model can be quickly constructed, and the fraud identification effect is improved.

This application discloses a network connection decision method, apparatus, device, and medium based on NAT state, applied to a network server and relating to the field of vehicle diagnostic technology. The method includes: obtaining a first message sent by a target device based on the monitoring results of a first socket and a second socket on the network server; if both the first socket and the second socket receive the same device serial number and socket number, determining whether the public network ports of the devices obtained by each socket are consistent, determining the target NAT type based on the determination result, and returning a response message to the target device within a preset time period; if both the first device and the second device meet the target information integrity condition, sending a second message to each device based on the target NAT type, so that the device can perform network connection operations based on the second message. This method improves the efficiency of UDP hole punching at low cost and quickly enables communication between the two devices.

An aerial target classification and recognition method based on an LSTM neural network is disclosed for classifying and recognizing target images. The method includes: Step S1, constructing a visualization module for analyzing target images; Step S2, inputting the original target image into the visualization module for recognition, and annotating the original target image to obtain a target image with annotated key parts, wherein the target image with annotated key parts contains several annotated key parts; Step S3, inputting the target image with annotated key parts into an LSTM neural network model for feature classification and recognition to obtain feature classification and recognition results; Step S4, fusing and analyzing all feature classification and recognition results to obtain moving target recognition results. This invention uses an LSTM neural network model to predict target types, focusing on key parts of the target in the image and eliminating interference from other objects, effectively improving target recognition accuracy.

The application discloses a kind of robot multimodal data intelligent transmission method and system, method includes: according to the event that robot accesses network under any broadcast network environment, by access gateway record multicast forwarding table item containing robot attribute information;According to the notification message sent by initiator robot to multicast communication group, carry target robot type, by multicast router robot attribute and target type in notification message are matched;According to the response message returned by target robot, determine unique target robot;According to the consistency judgment result of initiator robot and target robot access gateway, reaccess target robot gateway;According to the state that both sides are in same layer 2 network domain, directional data transmission not dependent on IP address is realized by exchange equipment.Utilize the embodiment of the application, the real-time and reliability of multi-robot collaborative system can be improved.

This invention discloses a barcode recognition method for pre-terminal junction boxes based on an Android application, belonging to the field of image recognition technology. It employs a YOLOv4 target detection network to detect 20 target types, including the pre-terminal junction box body, ports, and terminals, thereby significantly improving the accuracy of barcode area positioning. A relative position stability filter is used to match ports and barcode terminals, determining the correspondence between multiple ports and barcode terminals, ensuring that each inserted barcode terminal accurately finds its corresponding port number. Finally, a barcode recognition module is used to identify the barcode information. This method facilitates more convenient, faster, and more accurate fiber optic information entry and management for relevant personnel using highly flexible and portable handheld devices such as mobile phones, solving the problems of low efficiency, incorrect barcode information entry, and high operational difficulty in current manual fiber optic line maintenance.

This invention relates to a nonlinear node detector, comprising an RF signal processing module, a baseband signal processing module, and a logic control module. The baseband signal processing module is connected to both the RF signal processing module and the logic control module. This invention uses dual VCOs to isolate DC interference, which not only enhances signal quality but also improves the system's anti-interference capability and reduces the impact of DC components on mixer performance. By improving polyphase coding technology to generate an optimized fundamental frequency, the signal-to-noise ratio and detection accuracy are effectively improved. Combined with Kalman filtering to suppress noise and polyphase decoding to improve the harmonic signal-to-noise ratio, real-time spectrum analysis and target type identification are achieved. It can automatically select the frequency point with the least interference for operation, further improving anti-interference capability.

This invention relates to a vision-based method and system for identifying and classifying the boundary crossing behavior of intelligent agents in oilfield electronic fences. Belonging to the fields of oilfield perimeter security and computer vision technology, it addresses the problems of existing electronic fences, such as inability to identify target types, inability to quantify boundary crossing risks, single warning levels, and low system integration. The method involves target detection and continuous tracking of oilfield perimeter image data to obtain target type and movement trajectory. Boundary crossing verification is performed on the target's center position based on a polygonal electronic fence, and the boundary crossing distance from the target to the fence boundary is calculated. This distance is then fused with target type risk weights to obtain an adjusted boundary crossing distance. Classified warnings are implemented based on preset thresholds. Simultaneously, the warning information is visualized and logged, achieving accurate identification and differentiated warnings for boundary crossing behavior. This invention is applicable to perimeter security monitoring and intelligent risk warning in complex environments such as oilfield mining areas, well sites, and oil and gas storage areas.

Embodiments of the present application provide an object detection method and device, a storage medium and an electronic device, wherein the method comprises: in the case that a target sensor detects that an object of a target type appears in a target region, acquiring a real-time video stream obtained by a target shooting device shooting the target region; determining whether a moving target object appears in the target region in the real-time video stream, wherein the type of the target object is the target type; in the case that it is determined that the moving target object appears in the target region, sending a target signal, wherein the target signal is used to prompt that the moving target object appears in the target region. Through the embodiments of the present application, the problem of low accuracy of object detection in the related art is solved.