1348 results about "Detection performance" patented technology

The invention discloses an industrial Internet of Things time sequence self-supervision anomaly detection method and a monitoring and early warning system, and relates to the field of industrial Internet of Things, and the method comprises the steps: S1, constructing an anomaly detection model, and S2, obtaining a training data set; s3, training and optimizing an anomaly detection model; s4, acquiring to-be-detected data in real time; s5, performing anomaly detection analysis on the to-be-detected data, and outputting an anomaly detection result; through a time sequence and relation learning module, a dynamic graph topological structure learning module and an enhancement module, internal characteristics of a time sequence in a time domain and a space domain are deeply mined. The time sequence and relation learning module comprehensively captures a multi-scale time pattern, and the dynamic graph topological structure learning module eliminates dependence on a predefined graph structure; the enhancement module enhances the invariant representation under noise, and improves the recognition capability of the model to a normal mode; through wide experiments, the advancement of the method in detection performance is verified, and reliable support is provided for intelligent manufacturing and infrastructure diagnosis.

The invention discloses a pavement throwing detection method and device based on deep learning, electronic equipment and a program product. The method is realized through a trained detection model, the model adopts an LMSADet detection head, a multi-scale feature extraction and space attention mechanism is introduced into a task branch, and multi-scale modeling is decoupled from a backbone network and a neck network and integrated to the detection head so as to fit a detection task to directly optimize local details and scale differences of a throwing target. In order to suppress background interference and improve the recognition effect of fuzzy boundaries, the neck network is added into an MSHA module so as to efficiently capture the semantic relation between the thrown object and the background and enhance the regional understanding ability. A C3ESP module is introduced into the backbone network, deep features are extracted through stacking depth separable convolution, and information loss is avoided in combination with residual optimization fusion; meanwhile, a PEMA attention mechanism is introduced, the importance of different receptive field features is dynamically adjusted, the model focuses on key features, data information is captured more comprehensively, and therefore the detection performance is remarkably improved.

Systems and methods for dynamic invocation of synthetic probes based on Real User Monitoring (RUM) agents include monitoring application performance metrics using a Real User Monitoring (RUM) agent embedded within a client application, wherein the RUM agent continuously observes and reports metrics indicative of user experience; detecting performance anomalies by analyzing application and network metrics against baseline performance thresholds established during normal operations; and initiating dynamic synthetic probes in response to the detected anomalies, wherein said synthetic probes are adaptively configured to target relevant destinations, adjust probing frequency, and utilize specific probing methods tailored to the characteristics and severity of the performance anomalies.

The invention discloses a weak supervision video anomaly detection method based on prompt learning knowledge enhancement, and belongs to the technical field of video intelligent analysis. A video side gives a section of abnormal scene video, video sequence features and audio sequence features are obtained through a feature extraction network, then a trained and complete feature aggregation network is input to carry out multi-modal feature aggregation, an abnormal score is obtained through a score prediction network, and text representation is carried out based on prompt learning. A prompt template is constructed for abnormal video tags through a knowledge graph, semantic expansion is performed on normal tags through a plurality of learnable parameters, cross-modal alignment is performed on the normal tags and a video side, so that features of the video side are close to different normal semantics, knowledge enhancement is performed by introducing external information, positive abnormal boundaries of the video are learned, and the detection performance is improved. And finally, multi-task joint optimization is carried out through different loss functions, and abnormal video clip positioning is carried out.

The invention discloses a camouflage target detection method based on feature selection attention and frequency domain edge guidance. According to the method, four-level features of a camouflage target image are extracted through a backbone network SMT and are respectively screened; the high-level features are input into a semantic information supplement module, and after semantic features are enhanced, the high-level features and the trunk features are sent into a spatial feature enhancement module together. And inputting the obtained fine-grained features into an edge feature sensing module, and finally fusing multi-scale features through a multi-scale jump connection technology to generate a mask pattern with higher discrimination. The method has the advantages that the network parameter quantity is reduced and key information is reserved through a feature selection mechanism; a spatial feature enhancement module is used for enhancing multi-scale feature representation and remote dependence modeling; the dilution of the semantic context is relieved by means of a semantic supplement module so as to improve the positioning precision; and an edge feature enhancement module is adopted to enhance edge semantic perception and improve boundary integrity. According to the method, the camouflage target detection performance is remarkably improved with relatively low calculation cost.

The invention discloses a data-enhanced fine-grained multi-mode false information detection method and system, and aims to improve the accuracy of image-text false news detection. The method comprises the following steps: acquiring a news text and associated pictures thereof, extracting a text core entity semantic sequence and a visual entity semantic sequence by respectively utilizing a pre-training language model and a visual entity recognition model, and performing knowledge enhancement on original word-level text representation and low-level visual features through an attention mechanism; calculating a correlation matrix between the enhanced text and the visual features, dividing consistent and inconsistent regions, and respectively extracting consistent and inconsistent features; and finally, fusing the two types of features and global text representation to generate classification features so as to judge the authenticity of the news. Through double-channel knowledge enhancement and fine-grained cross-modal consistency analysis, the detection performance is remarkably improved, and the method is suitable for scenes such as social media and news platforms.

The invention discloses a camouflage target detection method and system based on a double-domain fusion enhanced network, and relates to the technical field of target detection. Through the nonlinear double-domain fusion module, in combination with nonlinear mapping of a spatial domain and a frequency domain, key difference characteristics of a frequency domain amplitude spectrum and a phase spectrum are captured, the problem that the detection performance is reduced in a scene of low contrast and the like depending on an RGB spatial domain is solved, and the target discrimination degree is improved; based on a lightweight scale perception modulation converter and a double-feature fusion module, multi-scale features are extracted, aligned and fused, a semantic relation is integrated by means of cross attention, and the problems of detail loss and boundary fuzziness caused by scale diversity are solved; the context feature enhancement module integrates cross attention and edge auxiliary injection, accumulates multi-layer feature integration, gives consideration to a global boundary and a local structure, effectively reduces false detection, missing detection and edge roughness, and further enhances robustness through multi-layer auxiliary supervision.

The invention relates to the technical field of deep learning, in particular to a photovoltaic cell defect detection method fusing multi-scale features and a re-parameterization strategy, and the method comprises the steps: obtaining a to-be-detected image; the image is input into a defect detection model, in the defect detection model, the backbone network is used for extracting a feature map of the image, and a first C3K2MSDA module is used for carrying out multi-scale cavity sliding window self-attention aggregation on the feature map to output a multi-scale feature map; the neck network is used for performing up-sampling on the multi-scale feature map and splicing the up-sampled feature map with the feature map from the backbone network to generate a fusion feature map, and performing multi-scale feature aggregation again by using a second C3K2MSDA module and an EMA-AFF module and generating a cross-scale fusion feature map; a DEC-Head module in the head network performs detection processing on the cross-scale fusion feature map, and outputs a defect category probability of each candidate region and a corresponding bounding box position as a detection result; and designing a loss function and optimizing the defect detection model by using the loss function. According to the method, the model defect detection performance can be improved.

The invention discloses a face forgery detection method and system based on multi-view collaborative fusion, and relates to the technical field of computer vision, and the method comprises the steps: obtaining to-be-detected video frame data; preprocessing the video frame data to be detected to obtain a standardized input image tensor; and inputting the standardized input image tensor into a pre-trained face counterfeiting detection model, and processing the standardized input image tensor by the face counterfeiting detection model to generate a face counterfeiting detection result. According to the method, the technical problem of poor detection performance of the model in cross-library and complex environments is solved, the detection robustness of the model in complex scenes such as fuzzy and compressed scenes is remarkably improved, and the detection precision is remarkably improved.

The invention provides a bearing ring surface defect detection method based on an improved YOLOv11 network. The method comprises the following steps: constructing an improved YOLOv11 network model; wherein in the backbone network and the neck network, an original standard convolution module of the YOLOv11 network architecture is replaced by a receptive wild coordinate attention convolution module; a surface detail fusion module is arranged on each of three feature map paths with different scales output from the neck network to the head network; a positioning loss function is configured to be a Focaler-DIOU loss function; training the model; and obtaining a to-be-detected bearing ring surface image, and inputting the to-be-detected bearing ring surface image into the trained model to obtain surface defect information. According to the method, the feature extraction capability of a network model on micro defects is remarkably improved, the detection performance of the network model on multi-scale and polymorphic defects is enhanced, and the positioning precision and convergence efficiency of the network model on irregular defects are improved.

The invention relates to the technical field of deep counterfeit image detection, and discloses a deep counterfeit detection method and system, a storage medium and computer equipment. The method comprises the following steps: firstly, constructing a reference data set containing a forged image and an original real image; secondly, through an integrated model, generating antagonistic samples for the reference data set, and integrating the successfully attacked antagonistic samples into an antagonistic sample set; and finally, merging the reference data set and the adversarial sample set, and constructing a robustness enhanced data set containing four types of samples. In the model training stage, multi-classification cross entropy loss and comparative learning loss are combined, and expression of the model in a feature space is optimized through comparative learning constraint, so that the model learns discriminative features with more compact intra-class features and more dispersed inter-class features. The model trained by the method not only can effectively defend against attack and improve robustness, but also surpasses original detection performance on clean samples, and has remarkable technical advantages and application value.

The invention belongs to the technical field of intelligent detection of new energy equipment, and particularly relates to a solar panel defect detection method and system, computer equipment and a storage medium, and the method comprises the following steps: a lightweight defect preliminary screening and adaptive shooting step: carrying out the recognition and image collection of a solar panel array through a lightweight model at the edge end of an unmanned aerial vehicle; a defect segmentation and type identification step: performing accurate segmentation on the solar panel and the defects through a neural network model, and judging the types and grades of the defects in combination with a feature extraction and classification model; a defect enhancement optimization step: enhancing the defects through an adversarial network; a detection performance evaluation step: quantitatively evaluating the overall performance of the system through a multi-dimensional index; mSAN-Net network segmentation is adopted, so that the defect detection precision is improved; and in combination with a GAN defect enhancement technology, the omission ratio and the false detection rate of weak defects are reduced, so that the fine operation and maintenance requirements of the solar panel are met.

The invention discloses a multi-pedestrian target detection method based on a YOLOCDG network in a complex scene, and aims to solve the problems of small target missing detection, model redundancy and inaccurate positioning in multi-pedestrian detection in a scene with dense people flow and serious shielding. The method comprises the following steps: firstly, based on YOLOv8, deleting a redundant convolution fusion layer of a backbone network, adjusting the size of a detection head, and simplifying a framework to reduce small target feature dilution; secondly, designing a context attention module CoAM and embedding the CoAM into a backbone network, and enhancing feature distinction degree in a shielding scene by capturing cross-target context association of dense pedestrians; thirdly, an improved C2fG module is proposed to replace a neck C2f module, the parameter quantity is reduced, and the detection performance and the edge deployment efficiency in a complex scene are balanced; then, an SPPFDSC multi-scale fusion layer containing depth separable convolution is designed, an original SPPF layer is replaced, and fine-grained feature perception of small-size pedestrians in the distance is enhanced; and finally, optimizing a bounding box loss function by adopting WIOU v3, improving the target positioning precision in the dense people stream, and finally constructing a multi-pedestrian detection model.

The invention relates to a balanced color perception enhancement method for rail transit target detection, and belongs to the technical field of rail transit and computer vision. According to the method, brightness channel adaptive histogram equalization enhancement CLAHE-LC, a multi-segment tone channel mask mechanism MSHCM-M and a three-stage hybrid mechanism non-maximum suppression TSLSH-NMS technology are combined, so that the problems of complex illumination, multi-target shielding, color interference and the like in a rail transit scene are solved, and the accuracy and recall rate of target detection are improved. The method can be seamlessly integrated into an existing deep learning target detection framework, is compatible with multi-label, multi-category and multi-scale features, and remarkably improves the precision and recall rate of target detection in a rail transit scene. Experimental verification shows that the real-time performance is guaranteed, meanwhile, the detection performance is obviously improved compared with a traditional scheme, and the method is suitable for being applied to an actual rail transit safety monitoring and intelligent maintenance system.

The invention discloses a privacy protection federated learning method and system based on Transform and knowledge distillation, and belongs to the technical field of artificial intelligence and network security, and the method comprises the steps: taking an attention mechanism of a Transform model as a core component of local feature extraction, so as to capture a data long-distance dependency relationship and improve the feature representation quality; a Paillier encryption protocol is introduced to realize homomorphic encryption transmission of model weights; the knowledge distillation technology is adopted at the central server side, the aggregation global model serves as a teacher model to extract soft knowledge, and the feedback client side compresses the model and optimizes the model; according to the invention, the Transform is used as a local feature extractor, the Paillier encryption protocol is combined, and the knowledge distillation technology is adopted after the central server is aggregated, so that the detection performance under Non-IID data is optimized, and both data security and detection efficiency are realized.

The invention provides an optical remote sensing image building detection method based on self-sensing dynamic multi-domain attention, and belongs to the technical field of remote sensing information processing and computer vision. The invention provides a remote sensing image building detection method based on self-sensing dynamic multi-domain attention to solve the problems that a fixed attention mechanism is difficult to adapt to multi-direction and multi-scale buildings and feature fusion is insufficient in an existing building extraction method. According to the method, a self-sensing dynamic angular domain attention module is used for enhancing multi-direction feature extraction through a self-adaptive rotation convolution kernel direction; the self-sensing dynamic scale domain attention module is used for dynamically selecting the size of a convolution kernel to be matched with buildings with different scales; and the self-sensing dynamic channel domain attention module realizes effective fusion of deep and shallow semantic features, and finally realizes end-to-end information extraction from input of a remote sensing image to a building detection result. The method has excellent detection performance in complex city scene remote sensing images, the IoU on an Inria data set reaches 91.95%, and the F1 score reaches 90.75%.

The invention discloses a covert communication optimization method based on deep reinforcement learning under limited character input, and belongs to the technical field of wireless communication. According to the method, an optimization problem which takes the minimum average bit error rate as a target function and takes hidden requirements, power and the like as constraints is constructed on a hidden communication model of a transmitting end, a receiving end, an intelligent reflecting surface and a monitor. Firstly, the bit error rate of a receiving end is given, the detection performance of a monitor is analyzed, and the KL divergence upper bound and the corresponding hidden constraint are deduced. The objective function and the constraint set of the constructed problem are non-convex, the optimization problem is trained and solved by constructing the optimization problem into a deep reinforcement learning model, the modulation order which is difficult to optimize by a traditional algorithm is optimized by using the advantages of deep reinforcement learning, a deep deterministic strategy gradient algorithm is selected to train the model, and then the error rate is minimized. And the covert communication performance is improved.

The invention provides a target detector construction method based on transfer learning in a foggy scene, and belongs to the technical field of deep learning. According to the method, two lightweight neural network modules, namely an image adaptive module and a feature attention enhancement module, are designed according to the requirements of an unmanned aerial vehicle target detector in a foggy scene, and the two lightweight neural network modules are integrated into a plug-and-play image defogging and enhancement network; and then designing a combined target detector network according to the characteristics of the image defogging and enhancing network and various widely used target detectors. According to the invention, the target detector can improve the target detection performance in a foggy scene under the condition that model parameters and computing resources are hardly increased, technical support is provided for reliable visual perception under a severe weather condition, and the method has a wide application prospect.

The invention relates to a small sample target detection method and device based on a decoupling prototype and a medium, belongs to the technical field of crossing of remote sensing image processing and computer vision, and can effectively separate interference features such as category semantics, angles and backgrounds of remote sensing targets. A double-branch prototype decoupling structure of'category semantic branch + angle geometric branch 'is constructed, and a decoupling loss constraint based on cosine similarity is introduced, so that a category prototype and an angle prototype are approximately orthogonal in a feature space; and then the RPN and the detection head are cooperatively guided through double prototypes, and a two-stage training strategy is combined, so that efficient migration of basic class knowledge to new class small samples is realized, and the small sample target detection performance and robustness in a complex remote sensing scene are improved.

The invention discloses an edge feature knowledge distillation-guided change detection method and device, computer equipment and a storage medium. Based on a feature knowledge distillation mode, hierarchical edge knowledge distillation is adopted, and an edge knowledge transmission channel from a teacher network to a student network is constructed. Through the improvement, the detection precision of the student network is obviously improved, meanwhile, compared with a fine and complex teacher network, the parameter quantity, the calculation quantity and the reasoning time of the student network are obviously reduced, and the network successfully achieves effective balance between fine detection performance and a lightweight network structure. The collaboration problem of edge learning and change detection tasks is solved. According to the method, effective balance is achieved among model refined edge perception, change detection performance and computing resource consumption, and reference and guidance are provided for practical application of an edge collaborative change detection network.

The invention provides a model compression and data enhancement fused lightweight deep forged voice detection method. The method comprises the steps of obtaining and processing a public voice data set and a large-scale self-supervision pre-training voice model; performing structured pruning and knowledge distillation to obtain a lightweight voice model; performing audio preprocessing and diversified data enhancement on the true and false voice samples to obtain an enhanced true and false voice data set; performing faking task joint fine tuning on the lightweight voice model to obtain a lightweight deep faking voice detection model; and locally deploying the model to obtain a localized counterfeit voice detection system, and carrying out real-time authenticity judgment. According to the method, calculation complexity and reasoning time delay are remarkably reduced, local deployment is carried out on a resource-limited end side platform, detection generalization and robustness are improved, and low-delay, low-power-consumption and high-robustness detection performance is achieved.

A computer-implemented system for foreign object detection in a scene. The system includes a memory-suppress diffusion network module adapted to reconstruct a reconstructed image from an encoded image, and a contrastive dissimilarity network adapted to combine the input image and the reconstructed image to predict an anomaly map for the input image. The encoded image is based on an input image, and the memory-suppress diffusion network module and the contrastive dissimilarity network are trained using only normal, real images. The system leverages only normal images in training and does not compromise the detection performance at the inference stage.

The invention discloses an audio deep forgery detection method and device, a terminal and a storage medium, and relates to the technical field of multimedia information security and artificial intelligence, and the method comprises the steps: constructing an audio deep forgery detection network model; according to the training set, performing decoupling stage training on an audio deep counterfeiting detection network model, and determining an initial training network model; performing meta-learning on the initial training network model according to the training set, and determining a target training network model; and obtaining a to-be-detected audio, inputting the to-be-detected audio into the target training network model, and determining an audio category corresponding to the to-be-detected audio. According to the method, meta-learning is adopted in the training stage of the audio deep forgery detection network model to force the audio deep forgery detection network model to learn general knowledge with higher generalization, so that the problem of poor detection performance caused by model overfitting in the face of an unknown vocoder and a natural scene in the prior art can be effectively solved.

The invention belongs to the technical field of digital image processing, and particularly relates to a knowledge distillation-based cross-modal fusion lightweight defect detection method, which comprises the following steps of S10, cross-modal fusion distillation; through a bidirectional vision-language alignment mechanism, the frozen multi-modal knowledge of a teacher model vision-language basic model VLM is migrated to a lightweight student model, and the dual-path fusion module comprises text condition region representation injected with semantic context and region anchoring semantic embedding fused with spatial vision clues; step S20, cross-header word-region alignment is carried out; embedding the fusion visual features generated by the two-way fusion module and the enhanced text to generate cross-head prediction so as to simulate the semantic-space association capability of a teacher model; s30, knowledge distillation loss is fused; according to the method, the multi-modal basic model is fused and distilled into the lightweight single-modal detection model, the detection performance in a defect detection scene can be improved, and compared with the basic model, the reasoning speed is greatly improved, and the parameter quantity is reduced.

The invention belongs to the technical field of automatic driving, and relates to a Transform attention mechanism-fused point cloud target detection method and system, and the method comprises the steps: 1, point cloud preprocessing: mapping an irregular sparse point set into a cylindrical voxel tensor of a fixed structure; 2, PillarVFE feature coding is carried out, and point-channel joint feature optimization is realized while the lightness of the network is maintained; 3, generating a pseudo image by the BEV backbone network, and mapping the pilar features to a two-dimensional aerial view plane after feature coding enhancement to form a dense BEV feature map; 4, outputting by a detection head, completing multi-scale feature extraction and context sampling fusion, and finally predicting a three-dimensional bounding box by the detection head; according to the method, the three-dimensional detection performance is remarkably improved in sparse, small-target and long-distance scenes while high reasoning efficiency is kept, and the method has good engineering practicability and research and popularization value.

The invention provides a low signal-to-noise ratio-oriented deep generative adversarial radar signal enhancement method and system. The method comprises the following steps of: extracting target scattering statistical characteristics from a preset historical database; generating an emission polarization state configuration instruction set according to the target scattering statistical characteristics; according to the emission polarization state configuration instruction set, obtaining an emission signal with a set polarization state parameter corresponding to the optimal emission polarization state, and emitting the emission signal in a low signal-to-noise ratio environment; receiving an echo signal generated after the signal is transmitted, and generating a noise polarization state parameter; obtaining a polarization filtering signal based on the noise polarization state parameter and a set polarization state parameter; based on target scattering statistical characteristics, obtaining a reconstructed polarization filtering signal, and obtaining an enhanced radar signal; according to the technical scheme provided by the invention, deep generative adversarial network reconstruction fusing optimal polarization emission, adaptive polarization filtering and target scattering characteristics is realized, and the radar echo signal quality and detection performance are improved in a low signal-to-noise ratio environment.

The invention discloses a vehicle-mounted track inspection system detection method, and relates to the technical field of tracks, an inspection starting point is set in a to-be-inspected line, a complete track image of the to-be-inspected line is pre-collected, a plurality of fastener images are generated through cutting with a fastener as the minimum unit, and track historical data are obtained; the track historical data are deployed on a cloud service side and a local inspection side at the same time; a vehicle-mounted track inspection system collects track real-time images, then the track real-time images are matched with an inspection starting point and cut into a plurality of fastener real-time images, region division is conducted on each fastener real-time image, track historical data of a local inspection side are called to conduct pixel grading comparison on different regions, and abnormal region images are screened; and incremental data is generated and transmitted to the cloud service side, track historical data is called to carry out refined detection on the abnormal region image, and a disease detection result is output. According to the invention, the problems of insufficient local detection performance, data accumulation and detection delay under high-speed operation of the electric passenger car are solved, and real-time track disease inspection is realized.

The invention discloses a remote sensing image target detection network based on hierarchical feature fusion and modal competition. The method comprises the steps of 1-3 for RGB remote sensing images and IR remote sensing images: 1, extracting low-layer local features of the remote sensing images by using a CNN (Convolutional Neural Network); performing global correlation calculation on the low-layer local features by using a Transform encoder, and outputting low-layer semantic features fused with long-distance spatial dependence; step 2, according to the low-level local features and the low-level semantic features obtained in the step 1, obtaining low-level difference enhanced low-level local and semantic features and channel enhanced low-level local and semantic features; 3, performing deep convolution fusion and feature abstraction on the difference enhancement and channel enhancement features of the same type to obtain low-layer compact local features and low-layer compact semantic features; the method provided by the invention has good detection performance on small target detection in a multi-mode remote sensing image and target identification in a complex view field environment while the detection precision and stability are maintained.

The invention relates to the field of computer vision and remote sensing image processing, and discloses a semi-supervised remote sensing target detection method, system, medium and equipment based on spatial resolution guidance, and the method comprises the steps: maintaining all types of GSD Gaussian distribution parameters in labeled and unlabeled data under a teacher-student framework; calculating category-level weights of the unlabeled images according to the matching degree of GSD distribution of the unlabeled images and GSD distribution of pseudo label categories in labeled data; calculating an image level weight based on the GSD distribution difference of the pseudo-label category between the labeled data and the unlabeled data; combining the supervised loss of the labeled data with the unsupervised loss of the double-weighted unlabeled data to serve as total loss; updating the student model through back propagation, and updating the teacher model through index moving average; and performing target detection by using the trained model. According to the method, pseudo label noise caused by resolution difference can be reduced, and stable training and detection performance improvement can be realized under a low labeling rate.

The invention discloses a power quality monitoring device adaptive detection system based on multi-source synchronous calibration and intelligent diagnosis, and belongs to the field of power system monitoring and metering detection. The system is composed of a multi-source signal calibration module, a dynamic time sequence alignment module, an intelligent error diagnosis module, a self-adaptive correction module and a data fusion calculation module. Through multi-source signal time mark synchronization, self-learning error correction and an intelligent diagnosis algorithm, high-precision and self-adaptive detection of the electric energy quality monitoring device is realized. The system can automatically identify and detect performance degradation and generate an early warning report, and has the advantages of high precision, high reliability and good engineering application prospect.