243results about How to "Improve inference speed" patented technology

The invention discloses a dynamic environment dense point cloud SLAM (Simultaneous Localization and Mapping) method and system based on YOLOv11 and ORB-SLAM 3. According to the method, real-time target detection and image segmentation technologies of a YOLOv11 model are integrated into an ORB-SLAM 3 framework, so that high-precision and robust visual SLAM in a dynamic environment is realized. A traditional convolution layer is replaced by using a balanced convolution method GSConv layer in a YOLOv11 model, and a traditional C2f layer is replaced by using a new feature fusion module VoVGSCS layer, so that a Neck structure of the YOLOv11 is improved, and a lightweight network model is realized. Experimental data prove that the pose estimation precision of the method in a dynamic environment is remarkably superior to that of an existing visual SLAM algorithm.

The invention belongs to the technical field of artificial intelligence and high-performance computing, and particularly relates to a method for improving the reasoning solution and training speed of a deep operator network model. According to the method, multi-level optimization is carried out on the PI-DeepONet model, and the reasoning solving speed and the training speed of the PI-DeepONet model are improved. According to the method, the SGEMM operator is customized, and compared with the best single-precision matrix multiplication in a basic linear algebraic library of a CUDA (Compute Unified Device Architecture) computing platform, the SGEMM operator can obtain a speed-up ratio of 1.5 times at most in a common model size range of PI-DeepONet. A fusion operator is designed according to the special structure of the modified MLP used in the model, the speed-up ratio of the end-to-end reasoning speed of the model is more than two times, and the calculation time of a graphics processor can be shortened by 5.7 times at most. According to the method, data parallel is used for carrying out distributed training on the model, a communication concealment strategy is used, and the expandability is kept at 83%-85%.

The invention relates to the field of artificial intelligence, and discloses a reasoning acceleration method and device for a large language model, equipment and a medium, and the method comprises the steps: obtaining a to-be-processed text in a reasoning request in response to the reasoning request input by a user; performing video memory demand prediction according to the to-be-processed text to obtain a video memory prediction result; distributing a corresponding initial video memory space for the large language model according to the video memory prediction result so as to execute a corresponding reasoning task on the to-be-processed text; when the large language model executes the reasoning task, monitoring an occupation index of the initial video memory space; and when the occupation index meets a preset condition, triggering a dynamic allocation mechanism, and allocating a new video memory space for the large language model according to the dynamic allocation mechanism. The corresponding initial video memory space is allocated by predicting the video memory demand in advance, the use condition of the initial video memory space is monitored in real time, and a dynamic allocation mechanism is triggered when resources are insufficient, so that flexible and controllable dynamic video memory allocation is realized, the waste of video memory resources is reduced, and the overall reasoning speed of the model is improved.

The invention discloses a vehicle driving track prediction method based on a large language model, and belongs to the technical field of automatic driving. The invention aims to provide a vehicle driving track prediction method based on a large language model, which is used for inputting a vehicle track into the large language model in a text format and evaluating the capability of the large language model in understanding the time dynamic aspect of a traffic agent. According to the method, tracks of different vehicles in a traffic scene are encoded by using a track encoder, track encoding features are mapped into a text embedding space of a large language model by using a reprogramming adapter, and the encoding features are translated into lexical combination which can be understood by the large language model. According to the method, powerful support is provided for realizing reliable and safe driving decisions by the large language model.

The invention provides a sewer internal defect detection method and system based on YOLOv11 improvement, and the method achieves the dynamic weight fusion of multi-level features through a self-adaptive feature fusion module, and improves the detection capability of a model for small targets and complex backgrounds. The selectivity of channel characteristics is improved through an SENet attention mechanism, and background interference is reduced; and introducing an MPDIOU loss function to improve bounding box regression loss. The performance of the improved model in detection tasks of sewer defects such as cracks, obstacles, corrosion, deformation and the like is remarkably superior to that of original YOLOv11, and particularly key indexes such as mAP50 (average precision), Box precision (P), recall rate (R) and the like are remarkably improved. According to the invention, efficient and accurate detection of multiple types of defects in the sewer can be realized, and a better solution is provided for maintenance and management of the sewer.

The invention discloses a lightweight metal plate surface defect detection method based on improved YOLOv5n, and relates to the field of machine vision and target detection, and the method comprises the following steps: S1, image preprocessing; s2, inputting the picture into the model; s3, model training; s4, data obtained after model training are analyzed and judged, the model is adjusted and optimized, and a final result is marked on an original image to serve as output of a detection result; and an improved YOLOv5n network model is adopted to carry out identification detection on the picture. The method has the advantages that the ADown module is introduced into the trunk network and the neck network to replace a CBS module in the original down-sampling process, the calculation efficiency is improved while the model parameter quantity is reduced, the deformable attention Transform is fused into the Bottleneck structure of the neck network C3 module, more feature information is obtained while the high reasoning speed is kept, and the method has the advantages of being high in reasoning speed, high in reasoning speed and high in reasoning speed. The method adapts to diversified defect characteristics of the surface of the metal plate, can meet the requirement for high detection precision, and meanwhile reduces the calculation amount of the model.

The invention discloses a lightweight bearing surface defect detection method and system based on improved YOLOv10n. The method comprises the following steps: acquiring a bearing surface defect sample data set; the method comprises the following steps: improving a YOLOv10 network structure to obtain an FM-NET algorithm model; training an FM-NET algorithm model based on the bearing surface defect sample data set to obtain a target FM-NET algorithm model; and performing defect detection on a to-be-detected bearing image according to the target FM-NET algorithm model, and determining a defect position and a corresponding defect category in the bearing image. According to the method, the multi-scale feature extraction capability and the model expression efficiency are remarkably improved, the robustness of the model in a complex scene is enhanced, the reasoning speed is remarkably improved on the premise of keeping high precision, and therefore good balance between the precision and the efficiency is achieved.

The invention relates to a remote sensing target detection method and device based on multiple scales and an efficient attention mechanism, and the method comprises the steps: obtaining a target remote sensing image, inputting the target remote sensing image into a target detection model, and obtaining a target detection result; wherein the target detection model comprises the following steps: replacing a 3 * 3 common convolution in the sixth network layer ELANBlock of the YOLOv7 by using a 3 * 3 depth separable convolution; replacing a 3 * 3 common convolution in the ELANBlock of the eighth network layer of the YOLOv7 by using a 5 * 5 depth separable convolution; using a 7 * 7 depth separable convolution to replace a 3 * 3 common convolution in the ELANBlock of the tenth network layer of the YOLOv7; an MIA-M module based on multi-scale efficient attention is used for replacing the 17th network layer ELANBlock and the 23rd network layer ELANBlock of the YOLOv7; an MIA-S module based on multi-scale efficient attention is used for replacing the ELANBlock of the 21st network layer of the YOLOv7; and replacing the ELANBlock of the 25th network layer of the YOLOv7 with a multi-scale efficient attention MIA-L module to obtain a target detection model.

The invention relates to a table question and answer method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a form file and question data uploaded by a target user; analyzing table information in the table file to obtain a candidate label set; based on the question data and the candidate tag set, determining a spoken language understanding result corresponding to the question data, wherein the spoken language understanding result comprises an intention and a keyword corresponding to the question data; performing reverse verification on the spoken language understanding result according to the question data to obtain a verification result; under the condition that the verification result indicates that the spoken language understanding result passes verification, a cue word is constructed according to table information, question data and the spoken language understanding result in the table file; calling the large language model according to the cue word for question-answering to obtain a question-answering result. By adopting the method, a large-data-volume table can be processed, and meanwhile, the accuracy and the stability of large-model table question answering can be improved.

The invention discloses an image restoration method and system based on gated convolution and a fused gated attention mechanism, and belongs to the field of image super-resolution restoration frameworks, and the method comprises the steps: taking a preprocessed image as the input of an image restoration model, carrying out the feature extraction and dimension raising, and obtaining a feature map; the image features are extracted by fusing the attention and self-attention mechanism of the convolutional neural network type, and iterative optimization is carried out by using the loss function, so that the local and global features of the image can be processed more effectively, the image reconstruction quality is improved, and the authenticity and detail richness of the high-resolution image are ensured at the same time.

The invention relates to the technical field of scene recognition, in particular to a real-time semantic segmentation method and system for a multi-shape pyramid in a traffic scene, and the method comprises the steps: initializing an original to-be-segmented image, and obtaining an initial segmentation feature map; channel splicing is carried out on the initial segmentation feature map and an output feature map obtained after first downsampling is carried out on the original to-be-segmented image, downsampling is carried out after a fused feature map is obtained, detail information is extracted, and a detail information feature map is obtained; down-sampling the detail information feature map and extracting semantic information; the method comprises the following steps: extracting transverse and longitudinal receptive field information of a semantic information feature map, performing expansion convolution to obtain feature maps of different receptive fields, obtaining a multi-scale and multi-shape context information feature map, performing up-sampling, performing point-by-point convolution dimension reduction on a detail information feature map, and performing channel splicing on the two features to obtain a fusion feature map containing semantic and detail information. And adjusting the channel number into a prediction category number, and normalizing to obtain a prediction segmentation result. The method has good segmentation precision and reasoning speed.

The invention discloses a lightweight crack penetration degree detection method based on ground penetrating radar and Swin Transform optimization, and the method comprises the following steps: 1, collecting ground penetrating radar data, and carrying out the preprocessing and feature extraction; step 2, constructing a multi-feature fusion data set; step 3, designing a two-stage through crack detection model based on Swin Transform optimization; 4, a first-stage through crack detection model optimized on the basis of Swin Transformer-YOLOv8 is designed, and a second-stage through crack detection model is designed on the basis of Swin Transformer-YOLOv8 optimization; and step 5, training and predicting the data of the training set and the test set according to the model architecture in the step 3 and the step 4. According to the invention, the accuracy of crack penetration degree detection is improved, and a scientific basis is provided for road maintenance; the detection speed is greatly improved, and the requirement of large-area rapid detection is met; the calculation complexity is reduced, the equipment performance pressure is reduced, and the detection efficiency is improved; the generalization ability of the model is enhanced, and the method can adapt to different road conditions; and the development of an intelligent road detection technology is promoted.

The invention discloses a model pruning method based on channel attention and adaptive learning, and the method comprises the steps: firstly introducing a channel attention mechanism SENet, adding the SENet to each convolution layer of a target model, and training the SENet until the model converges; then, extracting the middle output of the SENet of the training sample, combining with a scaling factor in a BN layer to obtain a channel importance score, and then performing pruning according to the channel importance score; and finally, training the pruned model by using an adaptive learning loss function until convergence. According to the method, the model pruning principle and the channel attention principle are fully utilized, and the high-precision model pruning method in the remote sensing field is achieved. The method can greatly reduce the consumption of computing resources, improve the processing efficiency and reduce the hardware cost while maintaining the model precision or even exceeding the model precision before pruning.

The invention relates to a question and answer method and device, computer equipment and a readable storage medium. The method comprises the following steps: inputting an initial question text of a user into a quantized language model to obtain word slot parameters corresponding to the initial question text; the word slot parameter comprises description information related to the target question and answer scene; the quantized language model is a language model obtained by compressing model parameters of the large language model; inputting the word slot parameters and the initial question text into the quantized language model to obtain a standardized question text; the standardized question text is a text which conforms to a preset text structure and clearly reflects the intention of the user; the standardized problem text is input into a vector model, and a plurality of documents with the highest correlation degree with the standardized problem text are obtained; and inputting the standardized question text and the plurality of documents into the quantized language model to obtain an answer text generated by answering the initial question text. The method can improve the reasoning speed of the question answering system.

The invention discloses a real-time traffic sign detection method based on deep learning. The method comprises the following steps: firstly, acquiring a traffic sign data set containing traffic sign images; secondly, constructing a traffic sign feature extraction network based on YOLOv8s, and performing feature extraction on the traffic sign image to obtain image features; and finally, constructing a lightweight shared detection head LSD, outputting a traffic sign detection result based on image features, and designing a loss function for reverse training. According to the method, the detection capability of the algorithm is improved, the scale change sensitivity of the network is weakened, more attention is paid to difficult samples, and accurate small target traffic sign detection is realized.

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 provides a student classroom behavior detection method based on Transform and task dynamic alignment. On one hand, the GLTB is added into the backbone network to enhance the feature extraction capability. Wherein the CGLU is used for enhancing the non-linear feature expression capability, and the global feature extraction capability of the MHSA is combined, so that the method has higher feature extraction capability. And on the other hand, a neck structure network improved by GLSA is adopted, global features can be effectively extracted, fine-grained local features can be captured, and interaction between global information and local information can be promoted. According to the method, when a complex task is processed, key information in the image can be identified and understood more accurately. In addition, a novel detection head ETADH is adopted, interactive features of classification and positioning tasks are learned through a feature extractor, and task decomposition is performed through a task decomposition module. In this way, the interactivity between the two tasks is enhanced, and the detection precision is improved by effectively utilizing the associated information between the two tasks.

The invention relates to the technical field of image detection, and particularly discloses a tongue image attitude correction and integrity detection method and system based on key point detection, and the method comprises the basic steps: firstly obtaining and processing a tongue image containing various kinds of information, constructing a neural network model for evaluation training, carrying out the transfer learning through a pre-training model, and optimizing the speed; the posture, integrity and accuracy are judged according to the positions of the tongue tip and the tongue root and the anatomical relation, a user is reminded of adjustment for detection abnormity, and if multiple failures occur, suggestions are automatically adjusted, and shooting is assisted. In the aspect of data processing, polymorphic tongue images are collected, enhanced and normalized, and the generalization ability of the model is enhanced. The network architecture is fused with ResNet and Transform, and the key points are accurately positioned. Performance is improved, real-time performance is guaranteed, accurate judgment is carried out according to the tongue tip position relation through scientific posture and integrity detection logic, and image quality is guaranteed. Clear prompt and auxiliary adjustment can be provided, tongue picture image acquisition is more efficient and accurate, and development of an intelligent traditional Chinese medicine tongue diagnosis technology is powerfully promoted.

The invention relates to the technical field of grassland ecological remote sensing intelligent detection, in particular to a toxic plant euphorbia fischeriana identification and spatial positioning method based on YOLOv8. The method comprises the following steps: optimizing a trunk feature extraction network, a neck network and a convolution module of a YOLOv8 network model to form an improved RFC-YOLOv8 network model; compared with the prior art, the method not only improves the detection precision of YOLOv8, but also enables the network model to be lighter, and improves the reasoning speed, thereby achieving the quick and accurate recognition and positioning of euphorbia fischeriana in a natural grassland, and finally achieving the spatial distribution drawing of euphorbia fischeriana.

The invention provides a near-electricity operation risk dynamic identification method fusing fine segmentation and binocular vision. The near-electricity operation risk dynamic identification method comprises the steps that a left view image, shot by a binocular camera, of substation near-electricity operation is acquired; initializing calibration parameters of the segmentation model, the binocular three-dimensional distance measurement model and the binocular camera; identifying and extracting key job elements in the left view image through a segmentation model, wherein a segmentation result comprises each detected instance and a two-dimensional position thereof; for each instance, the three-dimensional position of the instance in the space is predicted through a binocular three-dimensional ranging model, and the three-dimensional coordinates of the instance are obtained; according to the three-dimensional coordinates of each instance, the Euclidean distance between different key operation elements is calculated, and the minimum distance between the key operation elements is obtained; according to the invention, whether there is an operation risk is determined according to the minimum distance between different key operation elements, and the method has high distance measurement precision and real-time performance, and is helpful for improving the safety of near-electricity operation and the risk early warning capability.

The invention discloses a model early-leaving reasoning acceleration method, system and device and a storage medium. The method comprises: acquiring question and answer sample data of a large language model input by a user; based on an agent determination sub-model of a large language model, determining a first number of agents required for processing the question and answer sample data according to the question and answer sample data, and running a sample data reasoning sub-model of the large language model to reasone the question and answer sample data; an output result of each feature layer is extracted, and first accumulated time consumed when each feature layer outputs the result and a first memory occupied by the system when each feature layer outputs the result are determined; according to the first number, the first cumulative time and the first memory, determining a model early-leaving target layer number corresponding to each feature layer; and when the number of the feature layers continuously outputting the same result is the same as the number of the model early-leaving target layers, controlling the large language model to exit reasoning in advance and outputting the target result. The method can be widely applied to the technical field of natural language processing.

The invention discloses a multi-modal false information detection method, system and device and a storage medium, and belongs to the technical field of false information detection.The method comprises the steps that a text of multi-modal information is input into a text representation network to obtain text representation, and an image of the multi-modal information is input into an image representation network to obtain image representation; and inputting the text representation and the image representation into a shared classifier to obtain a text prediction information label, an image prediction information label, a text weight and an image weight, and weighting the text prediction information label and the image prediction information label based on the text weight and the image weight to obtain a prediction information label. According to the method, the text and image modal information is combined, and an adaptive weight distribution mechanism is adopted, so that efficient and accurate multi-modal information processing is realized, the prediction performance is remarkably improved, and the application range is widened.

The invention discloses a road surface disaster detection method and system, electronic equipment and a storage medium, and relates to the technical field of intelligent traffic management and computer vision, and the method comprises the steps: collecting image data containing a plurality of road surface disaster types, carrying out the data marking in each image data, and carrying out the data marking; a data sample is constructed through each piece of image data subjected to data annotation; training the improved YOLOv10 model by using the data sample until a loss function of the improved model reaches a preset end condition, and determining the improved model of which the loss function reaches the preset end condition as a disaster detection model; acquiring a to-be-detected image of a to-be-detected road surface, and inputting the to-be-detected image into the disaster detection model for processing to obtain a disaster detection result of the to-be-detected road surface, the disaster detection result including a disaster type and a disaster position; by combining the deep learning technology and optimization algorithm design, the problem that the existing method is insufficient in detection precision, robustness and applicability is solved.

The invention provides a DNN model-based division and reasoning acceleration method and system for vehicle-mounted edge intelligence. The method comprises the following steps: constructing a DAG graph based on a vehicle-mounted DNN model; constructing an auxiliary network flow graph of the DAG; adding a head node representing local calculation of a vehicle and a tail node representing calculation of an edge server or an auxiliary vehicle, and setting the weight of an edge as time delay overhead for segmenting the edge; obtaining different layering modes of an auxiliary network flow diagram of the DAG and corresponding maximum tolerance point sets meeting local time delay, memory and energy consumption limitation of the vehicle; correspondingly modifying the weight of the edge in the auxiliary network flow graph of the DAG based on each maximum tolerance point set, and calculating the maximum flow of each modified network flow auxiliary graph; and selecting a division mode corresponding to the maximum flow as a DNN model division mode. According to the method, the vehicle and road cloud cooperative reasoning is realized through a DNN model division mode, so that the DNN task cooperative reasoning completion time delay is reduced, and the vehicle safety and the user experience are improved.

The invention discloses an improved YOLOv8 target detection model construction method and application for small-size defects, and the method comprises the steps: constructing an improved YOLOv8 target detection model structure which comprises a Backbone module, a Neck module and a Head detection head; the Backbone module is formed by alternately connecting a plurality of CBS network structures and C2f network structures and finally connected with an SPPF network structure, and is used for carrying out feature extraction on an input image to obtain a plurality of effective feature layers; the Neck module is composed of a Conv network structure, a C2f network structure, an up-sampling network structure, a Concat network structure, a ConcatBiFPN and a Bi-Former attention mechanism network structure, bidirectional multi-scale feature fusion is carried out on an effective feature layer by using the ConcatBiFPN, and local spatial features are extracted and detail information is reserved by using the Bi-Former attention mechanism network structure; the Head detection head is used for outputting a target detection result; carrying out training and pruning processing on the built model structure, and deleting redundant parameters and neurons; the target detection model constructed by the method can perform high-precision detection on small-size defects, and the detection efficiency can be ensured.

The invention discloses a vehicle high beam detection method and system, and relates to the technical field of computer vision, a YOLOv8 network is improved by designing a feature fusion global attention module (CGM), a convolution feature recombination module (CCM) and a small target detection module (CSTC), and a vehicle high beam data set is adopted to train an improved YOLOv8 target detection model, so that the detection accuracy of a vehicle high beam is improved. A model adaptive to an edge computing hardware platform is obtained through format conversion and quantization; the core number matched with the hardware resource specification is adopted in an edge calculation hardware platform, collaborative reasoning is carried out on the quantized model based on night road vehicle images obtained in real time, and a real-time vehicle high beam detection result is rapidly obtained. Through targeted optimization of the architecture and deployment mode of the vehicle high beam detection model, the speed and performance of the model deployed in edge computing equipment are effectively improved, and real-time robust vehicle high beam detection is realized based on edge computing.

The invention discloses an information processing method and device based on a local large model, and belongs to the technical field of networks. According to the method, a task execution engine is constructed and deployed, so that a locally deployed large model can fully utilize the computing power of a graphics card, and efficient information processing is realized. Meanwhile, the video memory requirement is remarkably reduced through model quantification processing, and the reasoning speed is increased. A part of common system interaction related tasks are simplified by the task execution engine, so that a user can complete various tasks more conveniently, and the real-time response capability of the system is improved; based on a local large model, a user can perform intelligent dialogue and task execution even in a network-free environment, data privacy and security are guaranteed, and data privacy protection is realized; a user can initiate a task request without switching an interface, so that the operation efficiency and convenience are improved; the task execution engine can realize task execution deeply fused with the system, and the interaction between a user and the system is simplified; and more practical execution steps and methods are provided by disassembling complex tasks.

The invention discloses a data processing method and device, a storage medium and electronic equipment, and the method comprises the steps: obtaining first graph structure data corresponding to a target transaction, determining at least one transaction sub-graph structure data corresponding to the first graph structure data, determining a first transaction graph processing model based on the transaction sub-graph structure data, performing edge structure knowledge distillation processing on the first transaction graph processing model to obtain a second transaction graph processing model, obtaining second graph structure data of the target transaction, and performing graph sparsification processing on the second graph structure data based on the second transaction graph processing model to obtain target transaction graph structure data, and performing target transaction processing based on the target transaction graph structure data.

The invention discloses a text-based motion video generation method and device, a storage medium and equipment, and belongs to the technical field of computers. Obtaining a positive prompt word and a negative prompt word; performing text coding on the positive prompt words and the negative prompt words to obtain text features; the Animatediff model and the LCM Lora model are inserted into a Stable-Diffusion model, and a video generation model is obtained; processing the text features by using a video generation model to obtain a motion video; and performing video enhancement processing on the motion video by using the video enhancement processing model to obtain a high-definition motion video. According to the method, the LCM Lora model can be used for reasoning acceleration, the ControlNet model is used for enabling the video action to be consistent with the action in the action reference video, and the video enhancement processing model is used for generating the high-definition video.

The invention provides an improved convolutional neural network-based sky-bead image automatic identification method and system, and the method comprises the steps: collecting and preprocessing sky-bead images with different patterns and shapes, carrying out the data augmentation operation of the sky-bead images, and building a training set, a verification set and a test set which are needed by a sky-bead image identification model; constructing a convolutional neural network model based on improved OfficientNetV2, introducing a new activation function and an attention mechanism, loading training and verification sets in batches to carry out iterative training on a verification model, and continuously adjusting network parameters to obtain an optimal sky bead image recognition model; inputting a test sky bead image to the optimal model, and returning a sky bead image automatic identification result; according to the method provided by the invention, the accuracy and reasoning speed of sky bead image recognition can be improved on the premise of reducing calculation parameters, a lightweight model can be built for application, the problems of low accuracy, time and labor consumption and the like of traditional manual recognition are solved, and rapid and accurate recognition of the sky bead image can be realized.