365 results about "Category recognition" patented technology

The invention provides an underwater target identification method and system based on multi-source sensor data fusion. The method comprises the following steps: firstly, synchronously acquiring an original sound wave reflection signal and original optical image data of a target underwater area, then carrying out acoustic compensation processing on the original sound wave reflection signal to generate a target sound wave reflection signal, and meanwhile, carrying out optical compensation processing on the original optical image data to generate target optical image data; performing dynamic weighted fusion on the reliability measurement of the target sound wave reflection signal and the reliability measurement of the target optical image data through an adaptive fusion module to generate an enhanced multi-modal data set; and finally identifying the specific category of the underwater target through a multi-modal feature matching and decision-making mechanism. According to the technical scheme provided by the invention, the limitation of a single sensing mode is overcome, the quality of collected information is improved, the direct confirmation efficiency of common target recognition is also improved, and high accuracy and data reliability of target category recognition in a complex scene are ensured.

The invention discloses an emotion recognition method and system based on cross-modal alignment and matching enhancement. According to the method, firstly, feature extraction is carried out on text, audio and video modalities in a data set, and then a text and audio cross-modal emotion alignment module and a text and video cross-modal emotion alignment module are constructed respectively, so that cross-modal semantic alignment is realized. Constructing an emotion label matching module based on an alignment result, generating modal pairs with similar emotions but different labels by using a difficult negative sample mining strategy, and paying attention to cross-modal emotion consistency through a dichotomy task guide model; performing modal feature fusion on the three modals through a six-layer attention crossing mechanism, finally splicing feature vectors, inputting the spliced feature vectors into a long-sequence context fusion modeling module for deep modal fusion, and capturing cross-modal interaction information; and the fused features are sent to an emotion classification module, and a final emotion category recognition result is output.

The invention provides a WSI long-tail data identification method and device based on multi-modal distillation guidance and a readable storage medium thereof, and aims to solve the problem of low tail category identification accuracy caused by long-tail distribution in weak supervision multi-instance learning of a digital pathological total image (WSI). According to the method, performance improvement is realized through double-branch integrated multi-instance learning, text feature extraction and prompt generation and multi-modal distillation optimization. The double branches are trained by adopting balanced distribution and original long-tail distribution data respectively, an aggregator is shared, and prediction consistency is constrained; text features containing semantic prompts are generated to serve as supervision signals, and semantic alignment is achieved through image-text comparison loss; soft and hard label collaborative distillation loss is designed, and a total loss function is constructed in combination with consistency loss. Experiments show that the algorithm can significantly improve the tail category and overall recognition performance, is high in universality, can adapt to various multi-instance learning methods, and provides reliable technical support for intelligent analysis of pathological images.

The embodiment of the invention provides an image processing method and related equipment, and the method comprises the steps: obtaining a first sample image which comprises one or more objects; calling a pseudo tag generation model to perform feature extraction processing on the first sample image to obtain a first feature map; the pseudo label generation model is obtained by training a second sample image, and the image styles of the second sample image and the first sample image are different; performing domain generalization processing on the first feature map through a pseudo label generation model to obtain a domain generalization first feature map; the style information contained in the first feature map of the domain generalization is less than the style information contained in the first feature map; and through a pseudo label generation model, performing category identification processing according to the first feature map of the domain generalization to obtain a pseudo label of the first sample image, wherein the pseudo label is used for indicating the category to which the object in the first sample image belongs. According to the embodiment of the invention, the generation quality of the false label of the image can be improved.

The invention belongs to the technical field of industrial visual inspection, and provides an obstacle detection model training method and an obstacle detection method based on comparative learning and feature fusion. The training method comprises the steps of obtaining a point cloud sample set; extracting statistical and geometric feature vectors of each point cloud sample; pre-training the multi-scale feature fusion network by using the point cloud sample set based on comparative learning; constructing an obstacle detection network, wherein the obstacle detection network comprises a pre-trained multi-scale feature fusion network and a classifier; training an obstacle detection network by using the point cloud sample set to obtain an obstacle detection model; the multi-scale feature fusion network comprises M cascaded feature extraction modules; the cross-layer splicing module is used for splicing the statistical and geometric feature vectors of the point cloud samples and M feature matrixes; and a channel attention module and a point cloud feature projection head. According to the method, the precision of 3D point cloud obstacle category identification and the generalization ability of the classifier can be improved, and the model training convergence speed is high.

The invention discloses a ship category identification method based on incremental learning, and the method comprises the steps: carrying out the improvement of a training mechanism and a learning strategy of an original YOLO11 model, inputting a training set 1 into the original YOLO11 model for training, and obtaining the optimal weight of the model; the optimal weight of the model is loaded to the improved YOLO11 model, and an incremental target detection model is constructed through an EWC algorithm and a periodic data playback strategy; training an incremental target detection model by using the training set 2, and introducing a differential regularization strategy to enable the model to obtain the ability of inhibiting disastrous forgetting; and testing the incremental target detection model by using the test set, and continuously adjusting the regularization strength to obtain an optimal detection result. Incremental learning and target detection are combined, new ship categories are continuously learned through the model, meanwhile, stable detection performance of original ship categories is kept, and the problem of knowledge forgetting in the model iteration process is effectively solved.

The invention provides a power transmission line icing detection method and system, and relates to the technical field of intelligent control, and the method comprises the steps: eliminating the transmission interference of an optimization processing signal through a Manchester encoding protocol and a local cache verification algorithm in a channel switching process, and obtaining a clean transmission signal; and inputting the clean transmission signal into a pre-trained deep convolutional neural network, extracting ice layer space attributes through a plurality of convolutional layers and pooling layers, performing feature fusion through a full connection layer, calculating ice layer thickness distribution, identifying ice layer categories through a softmax classifier, and identifying the ice layer categories based on a thickness distribution diagram and an ice layer category identification result. And generating a detection result containing the ice layer thickness value and the spatial distribution attribute. According to the method, regional accurate analysis can be realized, and local feature omission caused by traditional overall analysis is avoided.

This disclosure provides systems and methods for controlling a vehicle based on a combination of high precision detection and high recall detection. The disclosed systems and methods can efficiently generate trajectories by reducing duplicate detection or duplicate calculation of objects or obstacles of common and known object types and objects or obstacles without class identification.

The invention provides an electromagnetic signal automatic modulation identification method and system during testing based on time-frequency fusion. The method comprises the steps of firstly constructing an electromagnetic signal data set; secondly, optimizing a kernel function to realize time-frequency feature modeling, and constructing a double-input path; constructing a deep neural network again, inputting the preprocessed signal sample into a time-domain branch for processing, analyzing a time-frequency spectrogram through the time-frequency branch, and introducing a channel attention mechanism to realize feature fusion; secondly, freezing a time-frequency branch, only training a time-domain branch, outputting a logit value from an input signal through a neural network, and smoothing logit distribution into probability distribution through a scaling variable; carrying out sample slicing processing on an input signal, and realizing convergence by fusing a prediction result; and finally, optimizing affine parameters of the batch normalization layer to complete identification of the electromagnetic signal modulation category. According to the method, the recognition accuracy of the model in a complex wireless environment is improved, so that the electromagnetic signal can still be stably and reliably recognized under the condition of low SNR (Signal to Noise Ratio).

The invention discloses a power quality composite disturbance detection classification and time positioning method based on MS-TCN + +. According to the method, composite disturbance category codes are designed for sampling points, an electric energy quality identification model comprising a prediction generation stage and a plurality of refining stages is constructed on the basis of MS-TCN + +, and multi-time-scale disturbance signal features are extracted by using double expansion layers in the prediction generation stage. And smooth loss is added in model training to avoid sudden change of categories of sampling points in a time sequence, and finally category identification and time positioning of disturbance signals are completed through a classification result of the sampling points, so that powerful support is provided for treatment of power quality disturbance.

Generate accurate answers to user questions. [Solution] The information processing system includes: a state information acquisition unit that acquires state information indicating the state of an object in question in response to the notification of question information, which is a question from a user regarding the object in question; a category identification unit that identifies a question category, which is the category of the question; a user support control unit that selects the state information that is classified into the question category identified by the category identification unit from the state information acquired by the state information acquisition unit, generates an answer request information by attaching the selected state information to the question information, requests a language model to generate an answer to the question using the answer request information, and acquires the answer generated in response to the request.

The invention provides a multi-dimensional enhanced open vocabulary video instance segmentation method, which comprises the following steps of: 1, proposing a novel open vocabulary segmentation thought through mathematical modeling and analysis; step 2, preprocessing a video input by a user, carrying out frame sampling, and realizing category-related segmentation in a Transform architecture by respectively strengthening interaction of category texts, image features and query vectors; step 3, adopting an instance-driven TopK time sequence matching strategy to improve the stability and accuracy of cross-frame matching; and 4, improving the category recognition capability through multi-scale feature fusion, and optimizing target classification. The open vocabulary video instance segmentation method can efficiently perform open vocabulary video instance segmentation, is widely applied to the fields of video monitoring, automatic driving, video indexing and the like, and promotes the development of video understanding and reasoning technologies.

A generator fault identification method, device, product and medium relate to the technical field of generator fault category identification. The method comprises the following steps: acquiring a vibration signal and a noise signal; performing multi-mode decomposition processing on the vibration signal and the noise signal to obtain a first sub-mode component and a second sub-mode component; determining a vibration signal sequence based on each first sub-mode component and the vibration signal; determining a noise signal sequence based on each second sub-mode component and the noise signal; determining a vibration characteristic matrix based on the vibration signal sequence; determining a noise feature matrix based on the noise signal sequence; determining a first recognition result and a first confidence coefficient based on the vibration feature matrix; determining a second recognition result and a second confidence coefficient based on the noise feature matrix; when the identification results are inconsistent, determining a fault identification result according to the confidence coefficient; and when the identification results are consistent, determining the first identification result as a fault identification result. The method is advantaged in that generator fault identification accuracy is improved.

The invention discloses a light-weight tea leaf tender shoot detection method and device for tea leaf picking, belongs to the technical field of image recognition and target detection, and aims to solve the problems of sensing and shielding of tender shoot targets for tea leaf picking. Comprising the steps of obtaining a to-be-detected image in a tea picking scene in real time; processing the to-be-detected image based on a pre-trained lightweight tender shoot detection model to obtain position information and a category identification result of the tender shoot small target; the training process of the lightweight tender shoot detection model comprises the following steps: acquiring an image data set containing a tea tender shoot target; performing data division on the image data set containing the tea tender shoot target to obtain a training set; and training a YOLO RGS detection model taking YOLOv8n as a basic network by using the training set to obtain a lightweight tea tender shoot detection model. The method can improve the perception level of tender shoot targets in a tea leaf picking scene, and provides efficient and stable visual support for intelligent equipment such as agricultural robots.

The invention provides an image recognition method and device based on semantics and an image recognition system. The method comprises the steps of obtaining a target image; extracting inter-class edges of the target object; determining an inter-class edge in the target image and an image region surrounded by the inter-class edge as a target object image, and identifying the target object image to obtain a class identification result of the target object; and sending the category identification result to a safety monitoring system of the financial institution, so that the safety monitoring system performs safety monitoring based on the category identification result. According to the scheme, the problem that in the prior art, the accuracy of identifying the target category of the image is low, so that the accuracy of safety monitoring of a financial institution is low is solved.

The embodiment of the invention discloses a multi-modal case database construction method and device, electronic equipment and a medium. A specific embodiment of the method comprises the following steps: performing layout region detection on a multi-modal medical case document to obtain a case region information set; performing hierarchical tree construction on the case region information set to obtain a case region logic hierarchical tree; generating case segmentation prompt word information; performing case segmentation processing on the multi-modal medical case document, and then performing structured information extraction processing to obtain a case text structured information set; performing category identification on the medical segmented case set to obtain a medical image category label information set; and performing multi-modal information association on the case text structured information set and the medical image category label information set to obtain a multi-modal synthetic medical case information set, and constructing a multi-modal case database. According to the embodiment, the multi-modal medical cases can be automatically extracted and integrated, the database construction efficiency is improved, and waste of storage resources is reduced.

The invention discloses an electronic material automatic classification method and system based on multistage feature recognition, and relates to the field of material classification, and the method comprises the steps: firstly obtaining unstructured material information, and carrying out the regularization and preprocessing of the information; subsequently, through main category identification and feature lexical element marking, accurate positioning of material core attributes is realized, which exceeds shallow identification of traditional keyword matching, and the defect that complex semantics cannot be processed is effectively overcome. On the basis, structured extraction and analysis are further carried out on the marked features, and the marked features are converted into paradigm features which can be deeply understood by a machine. And finally, through code integration and verification, a standard and unique production code, such as an SPYY. XBBBVVVV structure, is automatically generated. In this way, low-efficiency and error-prone manual operation is replaced, and the fundamental problem that an existing automatic scheme is difficult to deal with description diversity is solved through the powerful analysis capacity and robustness of the manual operation.

The invention discloses a target image recognition and target detection method based on a video enhancement algorithm, and relates to the technical field of image processing. The method comprises the following steps: firstly, receiving a rain, snow and fog scene video stream through a visual sensor, extracting a video frame target image, performing video enhancement processing, eliminating rain and snow shielding, fog blurring and noise, and generating an effectively enhanced image which is complete in target contour, clear in details and adaptive to subsequent detection; inputting the image into an improved YOLO model of the rain, snow and fog scene, completing feature extraction and category recognition through an optimized feature extraction network, outputting a preliminary target bounding box and a category label, and judging whether a target to be detected and a specific category exist or not; and finally, if the target exists, counting the detection data and carrying out validity verification, thereby realizing high precision, low misjudgment and strong real-time performance of target detection in severe weather of rain, snow and fog, and further effectively solving the problem of high detection result misjudgment rate caused by parameter adjustment lag of adaptive filtering in the prior art.

The invention discloses an AI marking closed-loop method based on active learning and difficult case mining, and the method comprises the steps: carrying out the parallel calculation of uncertainty, representativeness and diversity three-dimensional value indexes after training an initial model, and carrying out the dynamic weighting screening of a high-value sample through a closed-loop feedback controller; difficult cases are recognized, DBSCAN clustering is adopted, similar sample expansion is retrieved, and the samples are stored in a dynamic pool; annotations are distributed, intelligent judgment is triggered for dispute samples after consistency verification, and labels are determined by fusing node reputation, model confidence and feature similarity; a mixed loss function is adopted for incremental training, and if the performance does not reach the standard, sampling, difficult examples and training parameters are adjusted in a linkage mode and then retry is conducted; the effectiveness of the difficult cases is evaluated after each round of iteration, the invalid difficult cases are eliminated, redundant clusters are combined, and the pool capacity is adaptively maintained; and the performance of the monitoring model is iteratively optimized until the standard is reached. According to the method, the labeling cost is reduced, and the boundary sample and long tail category recognition capability is improved.

The invention discloses an intelligent solid waste monitoring method and system based on multi-source data, and the method comprises the steps: carrying out the multi-source data monitoring of a solid waste putting region, mapping the preprocessed initial multi-modal data to an event reference time axis, generating a time mapping function, and carrying out the data processing of the time mapping function; performing space-time alignment on the initial multi-modal data based on a time mapping function and the spatial index of the initial multi-modal data to obtain candidate multi-modal data, performing cross-modal feature extraction on the candidate multi-modal data to obtain a feature embedding vector of each modal data, and performing attention feature fusion on the feature embedding vector of each modal data to obtain an attention feature fusion model; according to the method, multi-modal fusion features are obtained, solid waste category identification monitoring is performed on a solid waste putting area based on the multi-modal fusion features, cross-modal event synchronization is realized, cross-modal feature fusion efficiency is improved, a significant moment is focused by taking an event as a center, redundant data processing is reduced, and the identification accuracy of the solid waste is improved.

The invention is suitable for the technical field of visual inspection, and provides a zero-sample industrial anomaly detection method and device based on cross-modal semantic alignment and a medium, the method comprises the following steps: obtaining an initial image set and an initial annotation file of a plurality of parts, constructing a training sample data set, comprising an image sample, a text sample, a mask sample and a label sample of each part; based on the training sample data set, a zero-sample industrial anomaly detection model is constructed and trained, and the model comprises a visual feature coding module, a semantic feature coding module, a cross-modal feature alignment and fusion module and a prediction module; obtaining a target image and target text description information of a to-be-detected part, and inputting the target image and the target text description information into the trained zero-sample industrial anomaly detection model to obtain a target mask and a target label; based on the target mask and the target label, the industrial anomaly detection of the to-be-detected part is realized, the part category identification and segmentation accuracy is improved, and the industrial anomaly detection efficiency and accuracy are further improved.

The invention discloses a user power consumption behavior prediction method based on big data aggregation analysis, and particularly relates to the technical field of power data processing. The method comprises the following steps: constructing a user power consumption behavior matrix by collecting multi-dimensional historical power consumption data of a user; fusing environmental factors such as air temperature, humidity and weather types to generate an enhanced feature map; extracting key behavior nodes of a user through a multi-scale graph convolutional network, and introducing a time attention mechanism to generate a time sensitive weight vector; generating a user individual prediction feature vector in combination with the behavior node and the time weight, and outputting a future power consumption behavior prediction sequence; when the predicted volatility exceeds a set threshold value, model reconstruction and weight updating are triggered; if the volatility is within a threshold value, behavior category identification and prediction result output are completed; the method improves the accuracy and stability of user behavior prediction, has the advantages of being high in period recognition capability, sensitive in fluctuation response and high in adaptive adjustment capability, and is suitable for intelligent power grid load management.

The mental illness recognition system based on the optical flow features collected by a visual sensor comprises a mental illness expert consultation video data preprocessing module, a facial stress unit extraction module, an optical flow change feature unit construction module, a classification model construction module and a patient mental illness category recognition module connected in sequence. The mental illness expert consultation video data preprocessing module feeds the patient facial picture to the facial stress unit extraction module. The facial stress unit extraction module feeds the optical flow calculation method and the patient facial picture sequence to the optical flow change feature unit construction module. The optical flow change feature unit construction module feeds the optical flow change feature unit to the classification model construction module and the patient mental illness category recognition module respectively. The classification model construction module feeds the classification model to the patient mental illness category recognition module. The mental illness and normal sample classification recognition is realized under the condition that the user facial video has good clarity.

This invention belongs to the field of service robot visual scene recognition technology, and provides a knowledge-enhanced method and system for object category recognition in multi-service robots. The method acquires scene images of the robot's operation, extracts visual information feature maps from these images, obtains preliminary predictions of target detection boxes and their object categories, and converts these visual information feature maps into scene information vectors. Trusted target label nodes from a pre-constructed knowledge graph are selected, and a matching vector is formed based on the converted scene information vector and target detection box information. The similarity between the matching vector and edges starting from the trusted target label nodes is calculated, and the query prediction result for the object category in the scene image is determined based on the similarity. Finally, the query prediction result for the object category in the scene image is fused with the preliminary prediction result to obtain the final object category recognition result.

This invention provides a method for fish passage behavior recognition based on a multimodal large model. First, imaging sonar image sequences and underwater optical video sequences of fish passing through fish passages or facilities are simultaneously acquired, and the multimodal data are aligned through time stamp synchronization and spatial calibration. Then, fish target detection, segmentation, and feature extraction are performed on the sonar image sequences and optical video sequences respectively to obtain the fish's spatial location, body length, water depth, motion state, morphological structure, and posture features. The obtained acoustic and optical features are input into a multimodal coding and fusion model based on the Transformer architecture, and a unified fish behavior representation vector is constructed through a cross-modal attention mechanism. During the training phase, model parameters are adjusted through a multi-task learning approach using cross-entropy loss, mean squared error loss, and contrastive learning loss, ultimately achieving fish passage behavior category recognition and prediction of individual and group passage difficulty scores.

The application discloses a slice detection method, medium, equipment and device, wherein the method comprises the following steps: acquiring a field of view image of a slice in a slice clamp to be detected, and inputting the field of view image into a first slice category recognition model pre-trained, so as to perform image recognition on the field of view image through the first slice category recognition model, and obtain a first slice type corresponding to the field of view image, wherein the first slice type comprises an abnormal field of view image and a non-abnormal field of view image; if the first slice type is the non-abnormal field of view image, inputting the field of view image into a second slice category recognition model, so as to output a second slice type corresponding to the field of view image through the second slice category recognition model, wherein the second slice type comprises a clear image and a motion blur image; the slice can be automatically detected, the misjudgment caused by invalid content on the slice is avoided, and the accuracy of the slice detection result is improved.

The application discloses a modulation category identification method and related device, electronic equipment and storage medium, wherein the modulation category identification method comprises: extracting a first feature map of a to-be-identified signal; wherein the first feature map comprises first sub-feature maps of a plurality of channels, and the first sub-feature maps of the plurality of channels have the same resolution; performing prediction on the first feature map based on a plurality of dimensions to obtain weight parameters of the first feature map in each dimension, and weighting the first feature map based on the weight parameters in each dimension to obtain a weighted feature map in each dimension; wherein the plurality of dimensions comprise at least one of a channel dimension and a spatial dimension; fusing at least the weighted feature map in each dimension to obtain a fused feature map; and classifying based on the fused feature map to obtain a modulation category of the to-be-identified signal. The above scheme can improve the accuracy and robustness of modulation category identification.

The invention mainly discloses a quality detection cooperative interaction suite group, which is used for performing defect detection on a continuous material conveyed by a roll-to-roll mechanism. Particularly, the quality detection cooperative interaction suite set further provides the following functions: (1) intuitively displaying the position of the detected flaw on the continuous material; (2) when the flaw detection result is misjudged, a user (a quality inspector) operates a tablet computer provided by the quality detection collaborative interaction suite group to delete the misjudged flaw detection result online; (3) when a flaw category identification error occurs in a flaw detection result, the user operates the tablet computer to correct the flaw category identified by the error online; and (4) when missing detection occurs, the user operates the tablet computer to add one or more images containing flaws into the database.

The invention relates to the technical field of fine-grained visual identification, in particular to a method for improving fine-grained visual identification precision under a low data volume, which comprises the following steps: acquiring an image with a low data volume scene characteristic, constructing a data set based on the image, preprocessing the data set, and dividing the preprocessed data set into a training set and a test set; a visual identification model is constructed, the visual identification model takes a ResNet-50 model as a backbone network, a segmented attention calculation unit is inserted between a Stage 4 of the ResNet-50 model and a global average pooling layer, and a loss function of the model is constructed based on cross entropy loss, divergence loss and multi-head consistency loss output by the segmented attention calculation unit; training the visual identification model by using the training set to obtain a trained visual identification model, the training including optimizing model parameters by using a loss function; and inputting the test set into the trained visual identification model, and outputting a fine-grained category identification result. The performance under the condition of low data volume is improved, and the key area positioning precision is improved.

Some embodiments use specialized machine learning models to generate computing system artifact names which reflect actions, states, conditions, or other aspects of artifact functionality. Artifact creation mechanisms such as method extraction mechanisms, test creation mechanisms, and template extraction mechanisms are enhanced with functionality by which they obtain and suggest meaningful generated names for new artifacts instead of merely prompting users with placeholder names. Specified artifact name formats and name styles are matched during name generation, thereby improving code maintainability and software development efficiency. Generated names are automatically and proactively derived from artifact source code by code summarization, conditional statement location, algorithm category recognition, or name format matching, for example. Generated names are also derived from natural language descriptions in comments and other documentation. Naming gaps left by autocompletion mechanisms are reduced. Inconsistencies between updated artifacts and their names are detected and remedied.