248results about How to "Guaranteed recognition accuracy" patented technology

The invention discloses a fan blade inner cavity defect detection method and system based on image recognition and deep learning. The method comprises the following steps: performing multi-angle and multi-position image acquisition on an inner cavity of a fan blade through imaging equipment; the collected fan blade inner cavity image data set is preprocessed; generating a primary defect sample by using the first-layer generative adversarial network model; generating a complex defect image by using the second-layer generative adversarial network model; a convolutional neural network model is used for training; inputting the preprocessed fan blade inner cavity image data set into the trained convolutional neural network model; generating a defect report based on a defect detection result; when a new defect type appears, the recognition capability of the new defect is continuously optimized. Compared with the existing single detection technology, the multi-network combination method disclosed by the invention has the advantage that the defect detection efficiency and the classification capability are remarkably improved.

The invention discloses a bird abnormal behavior recognition method based on radar monitoring data, belongs to the technical field of target detection, and solves the problem that target recognition precision is affected due to incomplete target single image data information acquired by a data acquisition module of an existing system. A bird sensing model performs correlation analysis on the normalized data set, whether a target object is a bird target object is judged, a target object path track is generated based on a target object recognition result and a path optimization algorithm, and the path track of the target object is analyzed and calculated in combination with a weak supervision bird detection algorithm to generate a path anomaly degree; the bird sensing model can analyze and recognize the bird flight path based on the target object recognition result, and judges and recognizes the bird abnormal behavior in combination with the weak supervision bird detection algorithm, so that the recognition precision of the target object is ensured, and the early warning capability of the bird abnormal behavior is improved.

The invention discloses a method for detecting an external load of an automobile emergency starting power supply, and relates to the technical field of automobile emergency starting, in the step S2, after an output circuit is closed, a system monitors dynamic current and load state changes, extracts related transmission data and calculates an abnormal coefficient, and once the abnormal coefficient exceeds a preset abnormal threshold value, the system is started; and the system sends a start stopping instruction in time, so that excessive loss of a storage battery or a starting power supply is further avoided, and the safety of equipment and a vehicle is effectively guaranteed. And S3 and S4, constructing a starting result recognition model by using the change coefficient and the abnormal coefficient, and fitting a judgment index after performing linear normalization processing on the data by virtue of a deep learning technology. By comparing the judgment index Pdzs with the qualified threshold value, the system can intelligently judge whether starting is successful or not, automatically control power output, avoid excessive current output and prolong the service life of equipment.

The invention discloses a construction site inspection method, inspection system and equipment and a nonvolatile storage medium, and relates to the technical field of intelligent construction, and the method specifically comprises the steps: carrying out the inspection planning according to the inspection requirements of a construction site; based on the inspection plan, executing an inspection task, and obtaining inspection data of a target area; and analyzing the inspection data by constructing a deep learning algorithm model, and carrying out risk event classification based on an analysis result. According to the invention, a closed-loop processing mechanism of data in a platform-nest-unmanned aerial vehicle-platform is driven through control of the unmanned aerial vehicle cruise by the intelligent construction transportation and management platform for construction engineering, and a mobile supervision mode for engineering quality and safety is innovated.

The invention relates to the technical field of information security, in particular to an abnormal network traffic detection method. Comprising the following steps: S1, obtaining to-be-detected network flow data, and performing preprocessing to form a to-be-detected data set; s2, constructing an abnormal network traffic detection model, specifically, S21, acquiring original abnormal network traffic data, and performing preprocessing to form an original data set; s22, constructing a teacher model, selecting a BiLSTM model as a basic structure of the teacher model, and training the BiLSTM model by using the original data set to obtain the teacher model; s23, constructing a student model, selecting a one-way LSTM model as the student model, and training the one-way LSTM model by adopting a knowledge distillation algorithm to obtain an abnormal network traffic detection model; s3, inputting a to-be-detected data set into the abnormal network traffic detection model, and outputting a detection result; and the detection speed and the operation efficiency are improved.

The invention discloses a coal gangue grading and sorting method and system based on LIBS (laser-induced breakdown spectroscopy) and image analysis, and relates to the technical field of coal gangue sorting, the method comprises the following steps: acquiring spectral data of each sample in a mixed coal flow comprising coal briquettes and coal gangue by using an LIBS technology; comparing the obtained spectral data with a pre-established coal briquette spectral database, preliminarily identifying coal briquettes and coal gangue according to a comparison result, and completing preliminary sorting; obtaining a visible light image of the separated mixed coal flow, preprocessing the image, inputting the preprocessed image into a coal gangue detection model based on an improved Mask RCNN, identifying and segmenting all coal gangues in the image, and obtaining the equivalent particle size of each coal gangue; and according to a preset particle size range, the coal gangue is divided into different grades, and the coal gangue in the mixed coal flow is classified and sorted. According to the invention, accurate identification of the coal gangue and grading sorting of the coal gangue can be realized.

The invention provides a track fastener state recognition method based on a three-section type parallel network architecture. The method comprises the steps that a track inspection image is acquired; constructing a convolutional neural network (CNN) backbone network, and performing multi-scale feature extraction on the image of the track fastener; performing convolution processing and up-sampling processing on the fastener features by using the local feature branches to obtain local fastener features; performing global average pooling and convolution processing on the fastener features by using the global feature branches; the method comprises the following steps of: processing the characteristics of the fastener by using a mask generation branch of a framework based on Vision Transform, and generating a prototype mask of the track fastener; local, global and mask fastener features are fused; performing non-maximum suppression and mask generation on the fastener feature map; and analyzing the health state of the track fastener. According to the method, the CNN backbone network and the Vision Transform are combined, a new three-section parallel network architecture is provided, accurate identification and state analysis of the track fastener are realized through instance segmentation, and the identification accuracy in a complex environment is improved.

The invention provides a target tree species remote sensing recognition method based on space-air-ground cooperation, which comprises the following steps: performing adjacent superpixel combination on segmented pattern spots after superpixel segmentation, and performing feature factor-based model optimization on a current recognition model of a target tree species based on combined pattern spot distribution and field distribution data, so as to obtain a target tree species remote sensing recognition model. Target tree species recognition is executed based on the updated new recognition model, and correct recognition pattern spots are obtained; executing edge matching detection on the correctly recognized pattern spots, adjusting the current merging scale based on an edge matching detection result to obtain an optimal recognition model, and further obtaining a recognition result of the target tree species based on the optimal recognition model; according to the method, the accuracy of the pattern spot boundary is effectively improved while the recognition precision of the target tree species is ensured, so that the matching degree of the boundary of the segmented pattern spot and the actual boundary of the target tree species is effectively improved.

The invention relates to the technical field of communication transmission, in particular to a power line carrier signal transmission method and device applied to a direct current lighting system, and the method comprises the steps: obtaining the current of each street lamp controlled by each concentrator in the direct current lighting system, constructing a current difference index and a noise interference index between any two adjacent street lamps, the voltage drop difference and the distance difference of any two adjacent street lamps are analyzed, and the signal influence index of a power line between any two adjacent street lamps is calculated by combining the control signal amplitude difference; according to the time domain difference and the frequency domain difference between a control signal issued by a concentrator and a control signal received by a terminal street lamp, an overall damage index is constructed by combining the control signal issued by the concentrator and the change of the timestamp of the control signal received by each street lamp, and the overall damage index is calculated based on a noise interference index, a signal influence index and the overall damage index. And obtaining the optimal routing forwarding path of each concentrator by using an optimization algorithm. The transmission effect of the carrier signal can be improved.

The invention relates to an occluded face recognition method based on FaceNet and an attention mechanism. The method comprises the steps of data set preprocessing, face recognition network construction, network training and testing. According to the method, an MTCNN backbone network is adopted to carry out face region detection and face key point detection, an attention mechanism is used to focus the network on an unshielded part, a CNN neural network is adopted to map a face image to a 128-dimensional Euclidean space, and the similarity degree of two persons is judged according to the Euclidean distance between two portraits. The invention aims to solve the problem of how to obtain the face features of the non-shielding face part, so that the performance and robustness of a face recognition task are improved, and the face detection effect in a complex shielding scene is improved.

The invention relates to the technical field of image processing and intelligent reading, in particular to a logistic regression-based PDF (Portable Document Format) file cross-page table merging and extracting method and system. The method comprises the following steps: firstly, loading a PDF file, analyzing line information in each page one by one, and constructing a transverse line set and a vertical line set of each page; calculating coordinates of intersection points of all the intersected transverse lines and vertical lines in the same page, constructing a four-dimensional matrix of each intersection point, restoring a table structure, and marking a table region page number and a context book; the four-dimensional matrix comprises four elements 0 or 1 in one-to-one correspondence with the intersection points in the up-down and left-right directions; traversing all the table areas of the PDF file, searching the table areas needing to be merged, and merging the table areas; and updating the combined table areas to a large-size page, updating the page number, and deleting the two combined table areas. The method comprises the steps of firstly analyzing lines, then transversely and vertically grouping and adjusting the lines, then merging the lines, and finally extracting intersection points according to the intersection condition of the lines; extraction of the table intersection points is guaranteed, contradictory errors possibly caused by simultaneous extraction of the lines and the intersection points are avoided, and improvement of accuracy and efficiency of subsequent table structure analysis is facilitated.

The invention discloses a pavement crack real-time detection and classification method adopting a digital twin technology, and belongs to the technical field of urban intelligent operation and maintenance. In order to detect and classify pavement cracks in real time, the method comprises the following steps: performing image quality enhancement processing on an acquired image matrix to obtain a quality-enhanced image matrix; carrying out image pixel updating on the quality-enhanced image matrix to obtain a pixel-updated image matrix, and storing the pixel-updated image matrix as a pixel-updated pavement image; all images including cracks, spalling, pit slots and upheaval diseases are selected and marked, and a data set for real-time pavement crack detection is obtained; a pavement crack rapid identification method integrating multi-model features is constructed; constructing a road digital twinborn model based on a pavement crack rapid identification method integrating multi-model features; and establishing a pavement crack classification method based on the obtained road digital twin model. The pertinence and effectiveness of road repair are improved, the service life of the road is prolonged, and the total cost of road maintenance is reduced.

The invention discloses a single-cell multi-omics gene regulation network reasoning method based on comparative learning, and belongs to the technical field of biological information, and the method comprises the following steps: obtaining single-cell RNA sequencing data and single-cell chromatin accessibility data, and carrying out preprocessing to obtain a prior network of multi-omics data; loading the prior regulation and control network into a graph structure, and converting the graph structure into a feature space matrix; calculating a Pearson's correlation coefficient of two nodes on each side in the graph structure, when the Pearson's correlation coefficient is greater than a preset threshold value, taking the Pearson's correlation coefficient as a node pair, and storing the node pair and the corresponding feature vector and hidden vector into a data set; the data set is divided into a plurality of batches to be randomly input into the comparison learning model, and the node type of each node, the pairing degree of node pairs and node feature reconstruction vectors are obtained; and extracting the mapping of two nodes in the node feature reconstruction vector in the feature space by adopting a regulation and control network interpreter, and calculating the structural entropy of the nodes based on the mapping so as to determine whether the nodes are key nodes or not.

The invention relates to an in-vehicle gesture recognition method and system based on a directional loudspeaker, and the method comprises the following steps: collecting gesture data through the directional loudspeaker, and extracting an audio clip containing a gesture, thereby obtaining a gesture waveform; preprocessing the gesture waveform; sTFT transformation is carried out on the preprocessed gesture waveform, the gesture waveform is converted into a spectrogram, and the spectrogram is evenly cut and then serves as input of a gesture recognition model; and gesture recognition is carried out by using a gesture recognition model constructed based on the ResNet network. Compared with the prior art, the method has the advantages of high recognition accuracy, high recognition speed and the like.

The invention discloses a load identification method and system. The method comprises the following steps: acquiring parameter data of a household appliance in steady-state operation; constructing a first composite feature based on the parameter data; and carrying out load identification according to the first composite feature in combination with a preset first load identification model. Through fusion of time domain and frequency domain features, operation features of different loads are comprehensively represented, and the identification precision is improved; forward and reverse features of load data can be effectively captured by adopting the BiLSTM model, and the processing capability of complex time series data is enhanced; a multi-head attention mechanism (MA) is introduced, a higher weight is given to key load features in the feature selection process, and the stability and generalization ability of the model are improved. The method is lightweight in design, is suitable for an edge computing environment, and can guarantee the recognition precision and reduce the computing overhead under the condition of limited resources.

The invention discloses a face recognition feature online updating and recognition method and device, and relates to the field of face recognition, and the method comprises the steps: matching a first feature of a face image with a first feature library corresponding to a lightweight model; similarity comparison is carried out on the second feature of the face picture and a second feature library corresponding to the high-precision model, and a first feature library corresponding to the updated lightweight model is judged through the similarity; the high-precision model is introduced on the basis of the lightweight model, and the first feature library corresponding to the lightweight model is updated through similarity judgment of the high-precision model, so that the accuracy and reliability of the features in the first feature library are improved, and online updating of the feature library in face recognition is realized. The recognition method based on the face recognition feature online updating method not only can guarantee the recognition rate, but also effectively improves the recognition precision, and achieves the high efficiency and accuracy of face recognition.

The invention discloses a network security threat perception device and method based on knowledge reasoning and algorithm service, and the device comprises a security threat perception and research and judgment subsystem which is configured to complete network security threat perception, and achieves the suspected threat research and judgment together with a knowledge extension subsystem, and generates a defense plan; the knowledge base knowledge extension and service subsystem is configured to extract knowledge from the network security threat field data for knowledge extension and provide knowledge service for the threat perception subsystem and knowledge base users; the intelligent algorithm model training and application subsystem is configured to perform intelligent enabling on a situation threat perception model uploaded by the threat perception subsystem, a knowledge reasoning model uploaded by the knowledge extension subsystem, a semantic association model uploaded by the semantic fusion subsystem and the like in the algorithm training subsystem; and the semantic knowledge fusion and application subsystem is configured to complete dynamic expansion, visual display and application service of the network security threat knowledge graph.

The invention belongs to the technical field of dust removal, and provides energy-saving type movable dust removal equipment and a working method thereof. The equipment comprises a controller, a camera, a laser detector and an adsorption assembly, wherein the adsorption assembly comprises an adsorption head and an adsorption pump; the controller is electrically connected with the camera, the laser detector and the adsorption assembly. Compared with a working mode of mobile dust removal equipment with a fixed adsorption rate in the prior art, the corresponding dynamic target adsorption rate can be determined according to the actual dust characteristics of the to-be-cleaned area, so that the adsorption rate of the mobile dust removal equipment can be dynamically adjusted, and the power consumption of the mobile dust removal equipment is effectively reduced.

The invention relates to the technical field of intelligent security and protection, and discloses an intelligent security and protection system based on AI face recognition, which comprises the following modules: an image acquisition module used for acquiring image information recorded at a camera of current security and protection equipment, a face detection and recognition module used for carrying out portrait feature recognition, and a face recognition module used for carrying out face recognition. The system comprises a portrait recognition module, an image acquisition module, an accompanying movement module, a data storage and management module, a human face recognition module, an accompanying movement module, a human face recognition module and an accompanying movement management module, wherein the accompanying movement module controls a video recording device in the image acquisition module to move horizontally and vertically; the face recognition module is used for establishing, managing and storing a database of face data. According to the invention, through the built accompanying moving module, when it is detected that there is an abnormal person in the current area, the AI controls the accompanying driving module to drive the image recording device to carry out accompanying movement of the angle and the recording position along with the movement track of the abnormal person, so that the complete abnormal behavior of the abnormal person can be traced more comprehensively.

The invention provides a diversified attack behavior recognition method and system based on a fusion recognition model, and the method comprises the steps: constructing an original training data set based on an attack received by a local end, and determining the type of missing training data based on the number of all types of labels in the original training data set; on the basis of a pre-trained data generation model, supplementing the training data of the missing training data category, and updating a local training data set; performing local training on the attack recognition model based on the updated local training data set, uploading model parameters of the trained local model to a cloud, and determining a model weight by the cloud based on the number of training data in the uploaded original training data set of the local model; and carrying out weighted calculation on the uploaded model parameters of each local model based on the model weights, issuing the updated model parameters to each local end to construct an aggregation model, and carrying out attack identification on data received by the local ends based on the aggregation model.

The invention relates to a special gas leakage detection method and system. The method comprises the following steps: acquiring an infrared image of a to-be-detected area; and inputting the infrared image into a pre-trained special gas leakage detection model to obtain a special gas leakage detection result. The method has the advantages that by proposing an infrared image enhancement algorithm based on MSRCR, WLS filtering and CLAHE, the image contrast and texture details are effectively improved, and the foreground and background marginal definition is enhanced, so that the robustness and accuracy of a subsequent deep learning network on leakage region identification are improved; a data set is expanded in various modes of network downloading, field shooting, software synthesis and the like, so that the model can learn richer features from a large number of diversified samples, the over-fitting risk is reduced, the generalization ability of the model is enhanced, and the model can adapt to different environments and leakage conditions; the combination of the lightweight model and the image enhancement algorithm enables the algorithm to achieve a faster detection speed while guaranteeing the recognition accuracy, and meets the real-time requirements.

The invention relates to the technical field of wearable intelligent equipment, and further relates to a method and device for determining the home position of a user and wearable intelligent equipment. The method comprises the following steps: when the wearable intelligent equipment is in an indoor environment, acquiring wireless signal data and multi-source sensing data; determining wireless signal feature data according to the wireless signal data, and determining a standing proportion, a charging behavior similarity and a position aggregation degree according to the multi-source sensing data; performing weighted summation on the wireless signal feature data, the standing proportion, the position aggregation degree and the charging behavior similarity; and when the weighted summation result is not less than the total threshold, determining that the target coordinate is the user home position. According to the method, the wireless signal data and the multi-source sensing data are effectively fused, the problem that a single judgment condition is prone to interference or misjudgment is avoided, the accuracy and reliability of user home position recognition are remarkably improved, a solid foundation is provided for subsequent intelligent services based on home scenes, and therefore the user experience is optimized.

This invention belongs to the field of materials testing technology and discloses a deep learning-based method and system for identifying the interface of SiC modified asphalt. It acquires multi-dimensional interface images through a layered time-series acquisition method using four devices: optical microscope, SEM, EDS-Mapping, and Raman spectrometer. Adaptive algorithms are used for targeted denoising, normalization, and registration, while attention mechanisms and weighted fusion strategies are combined to enhance interface features. Simultaneously, a multi-condition dataset containing a high proportion of minor defects is constructed to achieve deep fusion and accurate identification of multi-modal features, eliminating subjective errors from manual judgment and improving the fine-grained accuracy and detection efficiency of SiC modified asphalt interface identification. The U-Net model is improved for specific scenarios and designed to be lightweight. By optimizing the encoder-decoder structure, introducing a dual-channel attention mechanism and a condition-adaptive branch, and combining a loss function, the feature extraction capability of interface boundaries and minor defects is enhanced.

The embodiment of the present invention provides a control method, a mouse, a terminal and an Internet of Things system for an intelligent voice mouse, firstly collecting an original voice signal; then extracting its signal-to-noise ratio; finally, when the signal-to-noise ratio is greater than a preset threshold, according to a lightweight neural network, determining the target information in the original voice signal and transmitting it to a target terminal for identification; when the signal-to-noise ratio is not greater than the preset threshold, inputting the original voice signal into the lightweight neural network, obtaining the first output value of the noise processing module at the same time, inputting the first output value into the lightweight neural network, obtaining the target information and transmitting it to the target terminal for identification; wherein the input of the noise processing module is determined by the lightweight neural network and the original voice signal. Based on the original technology, the present invention adds a noise processing module to the terminal part, uses the intermediate value of the lightweight neural network to remove noise, and ensures the recognition accuracy under complex background noise.

This invention discloses a hybrid quantization method for image recognition models based on bit-flipping attacks, comprising: Step 1, performing initial quantization on the image recognition model to convert it into an initial quantized model; starting from the first layer of the initial quantized model, performing bit-flipping attacks layer by layer, and recording the sensitivity of each layer under attack; forming a sensitivity set of the initial quantized model from the sensitivities of all layers; constructing an optimization objective function and determining optimization constraints, and determining the optimal hybrid precision quantization strategy for the image recognition model based on the optimization objective function and the optimization constraints; Step 3, setting the bit width for storing the parameters of each layer of the image recognition model according to the optimal hybrid precision quantization strategy, thereby obtaining the optimal hybrid quantization model; Step 4, acquiring the image to be recognized and performing image recognition using the optimal hybrid quantization model.

The invention discloses a pavement pit slot repair material engineering quantity calculation method based on a digital twinborn model, and belongs to the technical field of pavement repair engineering. In order to accurately calculate the engineering quantity of the pavement pothole repair material, the method comprises the following steps: acquiring a pavement image by adopting a vehicle-mounted three-dimensional structured light camera, and then carrying out image pixel updating processing on the acquired pavement image to obtain a processed pavement image; extracting road surface images which all contain the road surface potholes and the disease types most similar to the image features of the road surface potholes; constructing a data set used for pavement pit slot identification; a lightweight pavement pit slot recognition method is constructed, and a deep learning model for pavement pit slot recognition is obtained; constructing a digital twinborn model of the deep learning model based on pavement pit slot recognition; and S5, based on the digital twinborn model obtained in the step S5, establishing a pavement pit slot repair material engineering quantity calculation method based on the digital twinborn model. According to the method, sufficient and accurate materials can be used for repairing work, and the repairing cost can be controlled.

This invention discloses a lightweight few-shot incremental learning method for power line defects. The method includes: first, constructing an inspection image dataset containing basic and incremental categories, and designing a lightweight feature extraction network; training the network using the basic category data to generate prototype vectors for each category, forming an initial prototype classification model; in the incremental stage, constructing incremental prototypes for a small number of samples of the new category, and introducing a prototype transfer mechanism based on inter-class tension, adaptively adjusting the new prototype based on the geometric similarity and discrimination conflict strength between the new and existing prototypes to alleviate category conflicts; simultaneously, introducing an adaptive feature fusion mechanism that perceives discrimination uncertainty, dynamically fusing the discrimination results of the main and auxiliary branches to improve recognition stability. This invention, employing the above-mentioned lightweight few-shot incremental learning method for power line defects, can achieve efficient and stable power line defect recognition in scenarios with few samples and continuously expanding categories.

The invention provides a substation drawing processing method and device based on deep learning, equipment and a medium. The method comprises the steps of receiving a transformer substation drawing and a design specification file which needs to be followed by the transformer substation drawing, dividing layout into a drawing area and a table area, processing the drawing area and the table area respectively, obtaining first structured data and second structured data, establishing semantic association of the first structured data and the second structured data, achieving association of the drawing area and the table area, obtaining structured semantic information, and displaying the structured semantic information. Analyzing the design specification file, and constructing a graph database knowledge graph and a vector database according to an analysis result; and searching and combining results by using structured semantic information, and judging whether the results accord with specifications or not. According to the invention, high-precision semantization and automatic compliance auditing of batch drawings can be realized, and the labor cost is reduced.

The invention discloses a high-precision radar target recognition system based on orthogonal waveform design and a convolutional neural network. The high-precision radar target recognition system comprises a radar orthogonal waveform design module, a waveform generation module, a transmitting antenna module, a receiving antenna module, a waveform receiving module and a convolutional neural network module, by designing an orthogonal radar transmitting pulse string, the receiving entropy is increased, so that a radar echo pulse string received by a receiving antenna contains enough target information, and strong difference is presented between a target and clutter as well as multipath interference characteristics; therefore, the convolutional neural network can extract target key features from a radar echo pulse string received by an antenna and suppress clutter and multipath interference at the same time. According to the method, radar echoes are directly processed in the electromagnetic original domain, features are efficiently extracted, high-precision recognition is achieved, traditional multi-step signal processing is integrated into one step, the process is simplified, the hardware complexity of a receiver can be reduced possibly, and the method has great significance in improving the performance of an electronic information system.

The invention discloses a copper rod surface treatment management and control method, system and device and a medium. The method comprises the following steps that a first monitoring video of a target copper rod conveyed to a first monitoring area based on a preset conveying mode is obtained; the preset conveying mode is that the target copper rod rotates while being linearly moved and conveyed; decomposing the first monitoring video frame by frame to obtain a plurality of first surface images of the target copper rod; according to the multiple first surface images, the oxide skin area of the surface of the target copper rod is obtained; whether the oxide skin area is larger than a preset first threshold value or not is judged, if yes, a pickling control program is obtained, and if not, a polishing control program is obtained; judging whether a preset condition is met or not, if not, returning to obtain a first monitoring video of the target copper rod conveyed to the first monitoring area based on the preset conveying mode, and if yes, ending; wherein the preset condition is that the oxide skin area is 0, and the method has the advantage that the better surface treatment mode can be flexibly selected according to the size of the oxide skin area.