129 results about "Relevant feature" patented technology

According to the movie and television play table book extraction method and device, the storage medium and the computer equipment provided by the invention, after an audio and video file of a movie and television play is split into a video file and an audio file, feature recognition is performed on the video file to obtain a subtitle text, speaker face information and a video understanding text; performing voice understanding on the audio file to obtain a voice transcription text and a voice understanding text; wherein the voice transcription text can be corrected into the standard transcription text with high accuracy through the subtitle text. Therefore, based on the face information of the speaker, the line segment of each speaker in the standard transcriptional text and the audio and video file is aligned, so that speaker information with accurate segmentation and semantic coherence can be obtained; and then, through combination with a character side-writing text generated by side-writing analysis on the speaker based on the video, the voice understanding text and the speaker information, table book information is constructed, and related feature description of the character can be covered on the basis of containing the line content, so that the content and depth of the table book are enriched.

A navigation path can be determined for an object using one or more neural networks. In various embodiments, image data is obtained that is representative of an environment in which the object is to be navigated. Relevant features are identified from the image, and a curve fit to those features. Loss values for the potential paths are scaled based at least in part upon the distance of those features in the real world. This can include, in at least some embodiments, performing the scaling as a function of the curvature of the curve fit to the features. Temporal smoothing can be performed with respect to prior path predictions in order to prevent sudden changes in the predicted path. The paths are analyzed to select a path with a highest confidence value that also at least satisfies a minimum confidence criterion. The path can be converted into three-dimensional navigation information.

The invention discloses an improved PointPill three-dimensional target detection method based on fusion sparse enhancement and geometric attention, and belongs to the field of deep learning and three-dimensional target detection. According to the method, an SP-Point Pill improved framework fusing a sparse Pill enhancement mechanism (SEPA) and a point awareness space attention mechanism (PASA) is provided. A sparse Pilllar enhancement mechanism is fused, and geometric information of sparse Pilllar is efficiently filled through neighborhood real point borrowing and a virtual point interpolation strategy; the point sensing space attention mechanism generates a dynamic weight based on the point cloud density and the average distance, and preferentially enhances target related features; the pedestrian detection method and the pedestrian detection device cooperate with each other to form a link closed loop of information completion and feature focusing, so that the pedestrian detection precision is remarkably improved on a KITTI data set, and meanwhile, the efficient reasoning speed is maintained.

The invention belongs to the technical field of computer vision and intelligent video analysis, and particularly relates to a multi-target tracking method based on trajectory recovery. The method comprises the following steps: firstly, performing target detection on an input video frame to obtain a bounding box, confidence and related feature information of a candidate target; then, a current detection result is matched with a historical track through Kalman filtering and appearance features, and track updating of the first stage is achieved; when an unmatched target exists, the position of the unmatched detection frame is dynamically corrected by calculating the average displacement of the center points of the front and back frame detection target, and whether the distance between the target and the camera is lower than the preset track recovery threshold value is judged by combining the distance information between the target and the camera. And when the conditions are met, the target is directly activated and recovered to be an effective tracking trajectory of the current frame, so that quick re-association of the lost target is realized.

The invention discloses an enterprise risk clue generation method and device, and the method comprises the steps: obtaining related feature data of an enterprise information system, inputting the related feature data into a trained risk clue prediction model, and outputting a risk clue prediction result of an enterprise; explaining the output risk clue prediction result through an SHAP method, and determining the contribution degree; and utilizing a causal forest model to predict the risk clues of the enterprise according to the related feature data and the corresponding contribution degrees, and obtaining an average value of output results of all the causal trees in the causal forest model as a final enterprise risk clue prediction result. Retrieving in an RAG knowledge database to obtain a corresponding text information retrieval result, and generating corresponding prompt word information; and inputting the prompt word information into the large language model to generate an answer to the enterprise risk clue. According to the invention, the accuracy and interpretability of risk clue prediction can be improved.

The invention discloses a depth image classification model evaluation method and system based on background pseudo-correlation measurement, and the method comprises the steps: obtaining a foreground target mask through a pre-training semantic segmentation model, and separating a foreground image; and generating various background images by using a text-to-image generation model in combination with random noise and semantic guidance weight. Then, background controllability constraint is adopted to adjust background change, meanwhile, the foreground is kept unchanged, and foreground and background images are fused to construct a composite image set; and then, inputting the synthesized images into a to-be-evaluated model, calculating category prediction probability difference, semantic representation of a feature extraction layer and an uncertainty metric value, and finally obtaining a correlation analysis error value to evaluate the dependency degree of the deep learning classification model on the pseudo-correlation features. Through combination of semantic segmentation, text-to-image generation and background controllability constraint technologies, quantification of the background pseudo-correlation dependency degree of the deep learning classification model is realized.

The invention relates to a food multi-modal detection data fusion analysis method and device, equipment and a medium. The method comprises the following steps: acquiring original multi-source heterogeneous data of a food spectrum, an image and a smell, and performing normalization and noise reduction preprocessing to obtain a standardized feature set; classifying and screening high-correlation features by using a support vector machine, and generating a structured description; dynamic parameters are extracted, a quality trend vector is generated through time sequence analysis, a safety score is calculated, and a qualified mark is output if the safety score reaches the standard; constructing an extension index set based on the score, inputting a dynamic model to generate an authentication update link, and combining the block chain time to obtain new-version authentication; and associating the historical records to generate an initial report, and verifying and correcting to obtain an optimized report. By adopting the method, the consistency and availability of food safety data can be improved, the credibility of safety certification is enhanced, and a systematic and efficient solution is provided for food quality safety detection.

The invention discloses an artificial intelligence model-based image classification robustness test enhancement method and system, and the method comprises the steps: carrying out the deep semantic analysis of the image content based on a multi-modal artificial intelligence model, automatically recognizing and extracting the class related features and non-class related features in an image, and generating a structured feature analysis report; the method comprises the following steps: generating a Keep Strategy and a Replace Strategy which are complementary to each other, and generating a Keep Strategy and a Replace Strategy which are complementary to each other; performing image editing operation based on the generated strategy to generate a corresponding image sample; performing automatic quality verification on the edited image sample, filtering out low-quality samples which do not conform to expectation, and generating a test sample; and comprehensively evaluating the target classification model based on the generated test samples, calculating performance indexes of the model on different types of test samples, generating a detailed robustness analysis report, and identifying weak links and improvement directions of the model. According to the scheme, a complete'generation-test-evaluation 'closed-loop system can be established, a standardized and quantifiable robustness evaluation index system is formed, and the model robustness can be scientifically and quantitatively evaluated.

This application provides a method and apparatus for auxiliary assessment of student quality based on a multimodal large model, belonging to the field of large model assessment technology. The method includes: acquiring student data of target personnel; using a multimodal large model to extract features from the student data in the following manner to obtain student quality features of the target personnel in multiple dimensions: converting structured data in the student data into corresponding student quality features according to preset mapping rules; extracting student quality features corresponding to unstructured text data in the student data using an attention mechanism based on preset prompts; and extracting student quality features corresponding to unstructured image data in the student data using a visual encoder. These multiple-dimensional student quality features are used to assist in assessing the student quality of the target personnel. This application can simultaneously process structured and unstructured data in the student data, extracting relevant features, facilitating the work of relevant personnel, and improving assessment efficiency.

The application provides a target tracking method and system fusing co-occurrence statistics and fhog gradient features, and the method comprises the following steps: determining an initial position according to a target region; calculating a target position p t and a scale factor s t‑1 of a previous frame of a current frame I t‑1 , so as to determine a target region and extract relevant features of the current frame; co-occurrence filter is used to obtain co-occurrence statistical information between pixel pairs in a tracked target image, fDSST algorithm is used to obtain fhog features of the tracking target, and the co-occurrence statistical information and the fhog features of the tracking target are fused, so as to be used as target image features; when the model is updated, the target region is determined according to a target center p t and a target scale s t of the current frame I t , and the target region is extracted; the target region extraction features are sent into a position filter and a scale filter, so as to obtain a target tracking result through iterative updating. The application solves the technical problems that the prior art cannot completely and effectively use feature expression of a target, has low robustness, and has poor tracking effect in a specific scene.

To provide an object identification method capable of monitoring an abnormal operation of a construction image in real time, saving labor costs required for safety monitoring at a construction site, improving an accuracy rate of the safety monitoring at the construction site, and reducing an occurrence rate of safety accidents.SOLUTION: The method includes: collecting a construction image; recognizing the construction image based on a target detection model to obtain a recognition result, the recognition result including a category of a construction object and location information of the construction object; and determining whether the construction image is associated with an abnormal action based on the recognition result. The ith feature segmentation module in the feature segmentation network performs convolution processing and segmentation processing on the ith type of first image feature, and outputs the jth type of sub-image feature to the jth feature fusion module connected to the ith feature segmentation module.SELECTED DRAWING: Figure 2

The application discloses a hyperspectral wetland image classification method based on a graph capsule neural network, which comprises the following steps: S1, learning feature transformation is performed on an adversarial domain self-adaptive framework, so that source domain samples and target domain samples of a hyperspectral wetland image are matched in features; S2, a graph capsule neural domain self-adaptive network structure is constructed, domain-invariant features and domain-related features are extracted, and transferable features are discovered and shared across domains; and S3, a coupling structure two-classifier is designed, the two-classifier is trained by using the source domain samples, classification differences of the target domain samples are maximized, and precise classification of the hyperspectral wetland image is realized by identifying a classification boundary. Meanwhile, the application discovers transferable knowledge and realizes cross-domain sharing, enhances effective discrimination of a class boundary, and finally realizes precise classification of the hyperspectral wetland image under conditions of unknown regions, complex scenes, and lack, deficiency and imbalance of data types.

The invention discloses a centrifugal fan running state intelligent diagnosis method and system based on deep learning, and relates to the technical field of centrifugal fan fault diagnosis and health management. According to the intelligent diagnosis method for the running state of the centrifugal fan based on deep learning, multi-working-condition vibration and multi-source data are collected and preprocessed, and weak fault features are enhanced through self-adaptive multi-scale time-frequency analysis; a parallel network is constructed, vibration deep features are extracted through physical prior attention, multi-source time sequence features are extracted through Transform, and the features are decomposed into fault sharing and working condition related features through decoupling loss and then fused; according to the method, field local feature distribution is aligned through local maximum mean value difference, and the data quality is improved through multi-source data synchronous acquisition and refined preprocessing. And by combining adaptive signal enhancement, physical prior guided feature extraction, feature decoupling fusion and local domain adaptation, working condition and domain difference interference is weakened, and accurate and stable diagnosis of the running state of the fan in a complex scene is realized.

The invention provides a semi-supervised quality monitoring method based on physical information guidance in a laser metal deposition manufacturing process. According to the method, firstly, an image processing algorithm is designed to extract relevant features (such as length, width and area) of a molten pool; secondly, analyzing feature relevance by using a Spearman correlation coefficient and calculating a feature weight; further, constructing a physical information gain loss function in combination with internal defect physical knowledge and feature weights; and finally, based on the physical information gain loss function, developing a total loss function for model training and optimization, thereby realizing quality grade monitoring. According to the method, the feasibility of monitoring the model based on the molten pool characteristic quality in the LDED manufacturing process is proved, the model performance is remarkably improved by integrating physics knowledge input, and the robustness and generalization ability of the model are enhanced.

The present disclosure relates to a font recognition method, device, readable medium and electronic equipment. The font recognition method divides the image to be recognized into a plurality of sub-images through the predetermined font recognition model, and obtains the first image corresponding to each sub-image. Image features, determine the second image features corresponding to the image to be recognized based on the first image features corresponding to each sub-image of the image to be recognized, the second image features include each sub-image of the image to be recognized and the context-related features of other sub-images are used to determine the font type corresponding to the target text based on the second image features. In this way, the image to be recognized can be described more comprehensively and accurately based on the correlation between each word image and other sub-images. This can effectively improve the accuracy of font recognition results and also effectively improve the font recognition rate.

The embodiment of the application provides a safety helmet detection method, a model training method, a system, a device and a storage medium, and belongs to the technical field of artificial intelligence. The second image feature is obtained by inputting the training data into the backbone network layer of the model for feature extraction and then connecting the network layer for feature integration. The input feature is subjected to average pooling, high-dimensional convolution and global average pooling operations in parallel in the connection network layer, and then the feature fusion is performed to obtain the output feature. Thus, the important local information and global context information in the image can be focused on, the loss of safety helmet feature information is reduced, the safety helmet detection related features in the comprehensive training data are learned, the second image feature is input into the head network layer of the model for feature prediction to obtain a predicted detection frame, and the safety helmet detection model is reversely updated based on the predicted detection frame and the real detection frame. The safety helmet detection model trained by the application has high robustness, and can accurately detect the safety helmet wearing condition for a blurred monitoring picture.

The invention relates to a method, a computer device, and a system for automatically detecting anomalies in machine operation. The machine operation in question is, in particular, that of machines for filling and packaging food and / or beverages. Anomaly detection comprises the acquisition of sensor data, the automatic categorization of this data according to operating states, and the extraction of relevant features for each category of operating states. Thresholds are determined using statistical methods to define precise operating limits. This model is monitored and adapted to react to anomalies at an early stage and to ensure operational safety. The invention provides a robust monitoring system that enables real-time monitoring of the machine system's condition and allows for early responses to deviations in operating conditions.This contributes to increased operational reliability, avoidance of downtime and optimization of maintenance processes.

Detecting application features within application portfolios. A processor may receive a request comprising an application. The processor may generate a first embedding vector based on the request. A model executed by the processor analyzes an embedding database that contains embedding vectors for multiple applications. Based on this analysis, the model identifies a first application within the plurality that features characteristics similar to those of the application in the request. Subsequently, the processor generates a report indicating that the first application includes these associated features and transmits the report to multiple recipients.

The invention provides a crab instance segmentation system and method based on multi-branch feature fusion. The method comprises the following steps: acquiring a high-resolution crab image in a unified illumination and fixed environment by using image acquisition; performing fine polygon segmentation marking on crab shells and step feet of the crab images, converting the crab shells and the step feet into a COCO format, and constructing a high-quality data set; a multi-branch feature fusion model is constructed in model training, shallow details, a local structure and high-level semantic information are respectively extracted by introducing a three-branch feature fusion path, and feature enhancement is performed on a weak texture region and a fuzzy boundary in combination with a feature enhancement module, so that the target boundary perception capability and the small target segmentation precision are improved, and the target segmentation accuracy is improved. Training to obtain a crab identification model; and performing instance segmentation on an acquired image by using the crab identification model, and outputting a target contour, a segmentation mask and related feature information. Therefore, high efficiency and cost advantages are achieved while high-precision instance segmentation is guaranteed, and the method has good practicability and popularization value.

The present method (300) identifies a speaker of interest in an audio file through a systematic approach. The process begins by receiving an input audio file via a processor (201). The audio file is then split into one or more chunks, followed by the extraction of relevant features from each chunk. Using a transformer encoder model, embeddings of the speaker of interest are generated based on these extracted features. The method identifies one or more nearest neighbours from various data structures corresponding to potential speakers, utilizing a classification model based on the generated embeddings. A set of nearest neighbours is then identified, ensuring that the count exceeds a predefined threshold and that the distance of each neighbour remains below a specified nearest-neighbour distance threshold. Finally, the method provides an identification of the speaker of interest as one of the recognized persons, enhancing speaker recognition capabilities in audio analysis.

The application provides a kind of drawing feature and digital matching method, system, equipment and medium, comprising: obtaining three-dimensional model and automatically extracting all relevant feature information;Obtain two-dimensional drawing and automatically extract all label information;Overall projection is carried out to three-dimensional model, and the model transformation projection image and projection information corresponding to three-dimensional model are generated;The model transformation projection image and two-dimensional drawing are carried out hash calculation, and the section projection picture position of the three-dimensional model corresponding to two-dimensional drawing is found;According to feature information, label information, two-dimensional drawing and step S3, determine three-dimensional model projection image and projection information, determine which part of feature of three-dimensional model corresponding to label information, and correspond label information to three-dimensional feature list;Output labeled three-dimensional model and feature label data details.The application can improve the efficiency and accuracy of two-dimensional drawing information recognition in production and processing, solve the problem of feature label information that two-dimensional drawing cannot be automatically recognized in automobile production.

The invention provides a text generation method and device, equipment and a storage medium, and relates to the field of data processing, and is used for eliminating a target tendency in text generation and improving the accuracy of text output, and the specific technical scheme comprises the following steps: mapping a to-be-processed text into a word embedding vector sequence, and constructing a paragraph matrix and a context matrix, global subject information and local detail information of the text can be captured respectively; performing attention feature extraction on the paragraph matrix and the context matrix to obtain a text feature vector fusing semantic association and target tendency related features, and identifying a potential target tendency in the text; target tendency related features are enhanced through context feature optimization, attention weights are dynamically adjusted on the basis of target tendency correlation scores, and spreading of target tendency information is inhibited in a targeted mode; and finally, based on the adjusted attention weight, generating a target text conforming to an objective standard, so that output misalignment caused by target tendency can be reduced, and the text output accuracy is improved.

The invention discloses a multi-industrial equipment-oriented ELM fault identification method fusing feature attention and manual LU decomposition, and belongs to the technical field of industrial equipment fault diagnosis. The core of the method is that a feature attention mechanism and a manual LU decomposition module are fused: firstly, the feature contribution degree is quantified through a channel attention mechanism, key fault features are automatically focused, weak correlation feature interference is inhibited, and the quality of a hidden layer output matrix is improved; secondly, the manual LU decomposition module realizes autonomous controllability and numerical stability of matrix inverse operation through row principal component selection and regularization optimization, and is separated from the dependence of a third-party math library. According to the method, the accuracy of equipment fault identification is remarkably improved, the omission ratio and the false alarm rate are effectively reduced, memory limitation of edge equipment is overcome, and an efficient and reliable real-time identification scheme is provided for bearing abrasion and gearbox abnormity of a wind power variable pitch system and faults of other industrial equipment such as a machine tool and an industrial motor.

The disclosure relates to a method, a computer apparatus and a system for automatically recognizing an anomaly in a machine operation. The machine operation is, in particular, of machines for filling and packaging food and / or beverages. Anomaly recognition comprises the capturing of sensor data, the automatic categorization of these data according to operating states and the extraction of relevant features for each category of operating states. Threshold values are specified with the aid of statistical methods in order to define precise operating limits. This model is monitored and adjusted in order to respond to anomalies at an early stage and ensure operational safety. The disclosure creates a robust monitoring system that makes it possible to monitor the condition of the machine system in real time and respond at an early stage to deviating operating conditions. This contributes to increasing operational safety, avoiding downtime and optimizing maintenance processes.

The invention discloses a facial expression recognition method based on feature deentanglement and self-distillation. The method comprises the following steps: S1, extracting an initial feature map from an input facial image through a backbone network; s2, generating a weight map based on the initial feature map, and decomposing the initial feature map into expression-related features and expression-independent features by using the weight map; s3, extracting supplementary features related to expressions from the expression irrelevant features, and fusing the supplementary features with the expression related features to obtain enhanced expression features; and S4, guiding the backbone network to carry out optimization training through a self-distillation mechanism by utilizing the enhanced expression features, so that the trained backbone network can directly output robust expression recognition features. According to the method, through characteristic decomposition, supplementary fusion and self-distillation optimization, the model is effectively separated and irrelevant interference information is inhibited, so that the accuracy and robustness of expression recognition are improved.

The embodiment of the present disclosure relates to a method for training a video recommendation model, comprising: obtaining a feature sample training set and a feature sample test set according to video-related features and user-related features; training a preset model according to the feature sample training set to obtain a prediction model, predicting the feature sample test set through the prediction model to obtain a first-class feature prediction index value set; predicting the feature sample test set through an evaluation model to obtain a second-class feature prediction index value set; comparing N second prediction index values in the second-class feature prediction index value set with corresponding first prediction index values in sequence, screening the second-class features with second prediction index values smaller than the corresponding first prediction index values in the second-class feature prediction index value set to form a feature set; training a video recommendation model by using the feature set, and determining feature identifiers of training features according to the feature set with the second prediction index values smaller than the first prediction index values; and improving the accuracy of the video recommendation model recommendation.

The embodiment of the invention discloses a scoring report information caching method and device, electronic equipment and a computer medium. A specific embodiment of the method comprises the following steps: sending database resource adjustment request information to a server; obtaining a user related data set; performing feature recognition on the user-related data set to generate a user-related feature information set; generating a related feature information set after score setting; generating a comprehensive evaluation information set; grouping the marked comprehensive evaluation information set to obtain a comprehensive evaluation information group set; grading and sorting the comprehensive evaluation information set to obtain a sorted comprehensive evaluation information set; generating a scoring report information set; and according to the scoring report information set, controlling an associated database server to carry out resource adjustment, and caching the scoring report information set to a preset memory area. According to the embodiment, the integrity and quality of the scoring report information are improved, the accuracy and safety of resource adjustment are improved, and resource waste is reduced.

The application belongs to the field of image processing, and discloses a knee joint image accurate automatic segmentation auxiliary method and system. The network structure is constructed, and the network structure is optimized according to the required result, so that the network structure is more matched with the knee joint image accurate automatic segmentation demand. Under the same data condition, more effective feature expression and more consistent segmentation output are obtained, and the dependence on single data distribution is reduced. The application obtains the source domain data set for training by processing the preprocessed image data, and trains the optimized network structure by using the source domain data set, so that the training data organization and the training process are more standardized and controllable, and the model is more likely to learn stable segmentation related features, thereby improving the adaptability and generalization ability to different source data, and solving the problems of insufficient model reliability under the conditions of insufficient adaptation of multi-center heterogeneous data and limited scale of high-quality labeling.

The invention relates to a cooperation capability evaluation method and device and a storage medium. The method comprises the steps of obtaining cooperation performance data of different test objects in a cooperation scene; wherein the collaborative performance data of the test object comprises face data, voice data, limb movement data, physiological signal data and eye movement data of the test object; for each test object in the different test objects, determining cooperation capability related characteristics of the test object according to the cooperation performance data of the test object; wherein the cooperative capability related features are features used for representing the cooperative capability of the test object; determining a collaboration capability evaluation result of the test object according to the collaboration capability related characteristics and a collaboration capability evaluation model; and outputting a collaboration capability evaluation result of the test object.