641 results about "Model compression" patented technology

The invention discloses an adaptive question-answering system based on knowledge distillation and multi-modal dynamic fusion, and the system comprises a knowledge distillation module which is used for migrating knowledge of a teacher model pre-trained on corpora in the communication field to a lightweight student model, achieving model compression through optimizing a distillation loss function, and obtaining a multi-modal dynamic fusion model; the loss function comprises a soft label output by the teacher model and a KL divergence constraint output by the student model; the multi-modal knowledge fusion module comprises a feature extraction unit, a self-adaptive weighting unit and an attention fusion unit; the self-adaptive inference engine comprises a semantic analysis unit; according to the cross-modal reasoning method and system, semantic alignment of equipment parameters, protocol texts and topological graphs is achieved through the multi-modal dynamic fusion technology, and the cross-modal reasoning accuracy is improved; compared with an original model, the lightweight student model has the advantage that the reasoning speed is increased in a protocol analysis task.

The invention discloses a field specialized large model fine tuning and deployment method and system based on adaptive optimization, belongs to the technical field of artificial intelligence and machine learning, and aims to solve the technical problem of how to improve the rapid adaptation capability of a large model on small-scale field data. In order to meet the requirements of special fields of law, medical treatment, finance and the like on professional terms and complex contexts, the technical scheme adopted by the invention comprises the following steps of: data processing and sample generation: performing cleaning, feature extraction and small sample expansion on field data, and generating a high-quality training sample through a field feature guide mechanism; small-sample fine tuning and migration: realizing rapid field adaptation of a large model through a small-sample learning technology and dynamic Prompt optimization, and reducing dependence on large-scale annotation data in combination with cross-field migration learning; compressing and optimizing the model; performing automatic deployment and collaborative reasoning; and monitoring and adaptive optimization are carried out.

The invention relates to the technical field of corn kernel quality detection, in particular to a corn kernel quality detection model construction method and system based on multi-source data fusion, and the method comprises the following steps: S1, multi-source data collection: collecting apparent morphology data of corn kernels through a multi-spectral imaging device, synchronously utilizing a near-infrared spectrometer to obtain spectral data of internal components of the corn kernels, and measuring structural density characteristic data of the corn kernels in combination with a sonic sensor; s2, multi-modal feature construction: forming a multi-modal feature matrix; s3, dynamic weighted fusion: generating a mixed feature vector; s4, constructing a dual-channel neural network: constructing a dual-channel deep neural network based on the generated mixed feature vector; and S5, model compression and deployment optimization: generating a lightweight detection model suitable for the embedded device. According to the invention, multiple requirements of an agricultural field on real-time performance, precision and deployment flexibility are met.

The invention discloses a road safety early warning method and system based on a mixed precision quantification visual large model, and the method comprises the steps: collecting road traffic safety videos and pictures, and carrying out the preprocessing, data enhancement and marking, thereby forming a diversified data set; a pre-trained visual large model is selected as a teacher model, after fine tuning, output layer and middle layer knowledge is extracted, key features are weighted, and meanwhile, a lightweight neural network is taken as a student model, same input is received, and prediction and middle feature maps are output. And inputting data into the two models and the student model to carry out mixing precision quantification forward propagation, constructing a total loss function containing tasks, knowledge distillation and quantification learning loss, and updating parameters through back propagation. And after training is completed, exporting a quantitative model, and deploying the quantitative model to an edge computing platform to realize safety early warning. The lightweight model can realize rapid reasoning on edge equipment such as a vehicle-mounted road side, and the problem that performance and efficiency are difficult to consider in a traditional model compression method is solved.

The invention relates to the technical field of security management, in particular to an AI-based intelligent security remote inspection and control method, which comprises the steps of front-end intelligent equipment deployment, intelligent inspection execution, data processing and intelligent analysis and remote control and linkage response. Compared with the defects of incomplete information and high false alarm rate due to the fact that equipment appearance anomaly detection mainly depends on single-modal data in the prior art, the scheme adopts multi-modal data preprocessing and feature extraction, constructs a cross-modal alignment network architecture, dynamically fuses laser radar, RGB-D images and IMU data by using a self-attention mechanism, and improves the detection accuracy of the equipment appearance anomaly. According to the method, more accurate equipment appearance abnormity identification is realized by combining environment self-adaptive adjustment, and the model is compressed and deployed to an edge end through a knowledge distillation technology, so that real-time reasoning and analysis are realized, the intelligent level of security inspection and the accuracy of abnormity detection are remarkably improved, the false alarm rate is reduced, and the practicability and reliability of the system are enhanced.

The invention provides a personalized fitness system and method based on multi-modal data and an adaptive large model. According to the method, multi-modal data are integrated, and heterogeneous data collaborative analysis is realized by using a space-time alignment algorithm and a confidence coefficient weighting mechanism. A multi-task learning framework is adopted, physiological prediction, exercise risk early warning and psychological incentive strategy generation tasks are synchronously processed, a user stage target is adapted through a dynamic attention mechanism, and federal learning and transfer learning technologies are combined. A nonlinear periodic planning engine based on reinforcement learning dynamically adjusts training load and action difficulty according to real-time biomechanical simulation results and recovery state evaluation, and meanwhile, a lightweight model compression technology and end-side reverse dynamics calculation are adopted. Based on the scheme, the exercise performance prediction precision and the early warning timeliness are improved, and the exercise loss risk caused by overtraining is reduced; the cold start data bottleneck is broken through; low-delay virtual coach interaction and privacy protection are realized; and dynamically adjusting the training intensity and the incentive strategy.

The invention provides a big language model pruning method and device, a storage medium and a program product. The method comprises the steps of obtaining a pre-trained big language model and a corresponding text data set; dependent neurons are activated through iteration, interdependent coupling groups in the large language model are recognized, and the coupling groups serve as basic units of pruning; based on the coupling group, performing coarse-grained evaluation and fine-grained evaluation on the parameters in the group, and fusing and calculating the importance of the parameters to generate a pruning mask; redundant parameters in the model are removed through pruning operation according to the pruning mask; performing parameter optimization on the pruned model, and performing fine tuning through a text data set to generate a target large language model; and deploying the target large language model to a computing device, executing a text generation or semantic analysis task, and counting the task accuracy rate and the model compression rate. According to the method, the parameter quantity of the model is effectively reduced, so that deployment is simple, and the downstream task processing accuracy of the pruned model is improved.

The invention relates to the technical field of artificial intelligence and edge computing, in particular to an end-side cloud collaboration system based on a hybrid scheduling strategy. The end-side cloud cooperation system based on the hybrid scheduling strategy comprises a model segmentation module, a state sensing module, a reasoning scheduling module, a model compression and deployment module, an edge optimization engine and a communication synchronization module. According to the end-side cloud cooperation system based on the hybrid scheduling strategy, elastic deployment and dynamic reasoning of a complex model in a multilayer heterogeneous environment are supported, the system state can be sensed in real time, a scheduling path can be optimized, and the robustness of the system is improved; and in combination with model compression and edge operator optimization, the reasoning efficiency is remarkably improved, the communication load is reduced, and then low-delay cooperation between the terminal and the cloud is ensured.

The invention discloses an optical storage charging station cross-site heterogeneous data fusion health state early warning method, which adopts transfer learning and domain adaptation technologies, measures the data distribution difference between a source site and a target site through the maximum mean value difference, realizes knowledge transfer and improves the generalization ability of a model. And on the basis of an Attention-LSTM feature fusion mechanism, feature weights are adaptively allocated, and the fault feature extraction capability is enhanced. And designing a fault mode sharing module, coding the fault mode of the source site into a knowledge base, and rapidly matching similar faults of the target site. A model compression technology is introduced, lightweight deployment is realized through knowledge distillation and a parameter quantification method, and operation of edge equipment is effectively supported. And furthermore, an online learning mechanism is adopted to dynamically update model parameters to ensure that the model adapts to the environment change of the target site. According to the method, the accuracy and robustness of cross-site fault prediction are remarkably improved, the model deployment time is shortened, and the method is widely applied to intelligent operation and maintenance of the optical storage charging and discharging integrated power station.

The invention discloses a multi-unmanned aerial vehicle and unmanned ship formation control method and device based on a graph convolutional network and deep reinforcement learning. The method comprises the following steps: constructing a first simulation environment of multi-agent formation; collecting motion control data of the unmanned ship according to the first simulation environment, and constructing a depth control model based on the motion control data and a sequence generation model; constructing a graph convolutional network model, and obtaining compression features according to the graph convolutional network model and the multiple agents; constructing an executor network and an evaluator network based on an independent near-end strategy optimization algorithm; obtaining an initial control model according to the graph convolutional network model, the compression features, the executor network and the evaluator network; and training and testing the initial control model in a first simulation environment according to the depth control model to obtain a target control model. The method can improve the sensing data processing efficiency and the real-time decision response performance, and can be widely applied to the technical field of multi-agent control.

The invention relates to the technical field of industrial intelligent inspection driven by artificial intelligence, and discloses a multi-modal property management scene risk point detection system based on artificial intelligence, and the system comprises a multi-modal data collection module which is used for synchronously obtaining visible light images, infrared thermal imaging and Internet of Things sensor data in building facilities; the domain self-adaptive defect generation module is based on a decoupling type generator and a StyleGAN2-ADA framework; a multi-modal feature fusion module; an attention enhancement detection module; a lightweight model compression frame; a real-time detection module; and a continuous learning module. High-fidelity defect samples are generated through the decoupling generator and the StyleGAN2-ADA, the problems of sample scarcity and class imbalance are solved, samples are generated through the cyclic generative adversarial network to cover multiple defect types, the acquisition period of the defect samples is shortened, and the requirement for training real samples is reduced.

The invention discloses a digital twin-driven multi-source heterogeneous data fusion and intelligent analysis method for oil and gas pipeline construction. The method comprises the following steps: collecting multi-source heterogeneous data related to oil and gas pipeline construction; fusing data by using a space-time diagram neural network and constructing a dynamic offset compensation model to correct coordinate drift; constructing a data credibility dynamic evaluation chain, performing cross validation on sensor data to generate a confidence coefficient weight, and complementing missing data through a GAN complementing module to form a credibility enhanced data set; designing a layered distillation compression algorithm to compress the PINN model into a TinyLSTM engine, and developing a model increment updating protocol to realize efficient collaboration of a cloud end and an edge end; and training and optimizing a TinyLSTM engine by using the data set with enhanced credibility. According to the method, the defects in the aspects of data dynamic correction, credibility evaluation and complementation, model edge deployment real-time response and the like in the prior art are overcome, and the method has remarkable innovativeness and practicability.

The invention relates to the field of computer vision and deep learning, and discloses an image big data classification and recognition method and system based on deep learning, and the method comprises the steps: generating a gating matrix through the extraction of an image frequency domain energy coefficient, compressing a convolution kernel through the combination of asymmetric tensor decomposition, and carrying out the self-adaptive training through cross-modal semantic alignment and a meta-learning task. Efficient classification reasoning of dynamic path selection is realized, and the precision and the calculation efficiency are improved; the system comprises a frequency domain analysis module, a dynamic sparse gating module, an asymmetric tensor decomposition module, a meta-learning task generation module, a cross-modal alignment module and a dynamic inference engine module. According to the method, through cross-modal semantic alignment and meta-learning task optimization, in combination with lightweight parameter storage and edge calculation path selection, fine-grained classification precision improvement, model compression and high-efficiency reasoning are realized, and the calculation efficiency and generalization ability in a complex scene are remarkably enhanced.

The invention discloses a hydraulic power plant automatic debugging system with a self-adaptive communication function. The system comprises a processing unit, a storage unit, a communication module, an input / output interface, a task classification module, a model architecture configuration module and other functional modules. The use method comprises the following steps: analyzing task complexity and high-dimensional features of input data through a task classification module, and grading by adopting algorithms such as principal component analysis and a support vector machine; the model architecture configuration module is used for matching a model architecture according to the complexity level and determining an initial calculation path; the dynamic path adjustment module optimizes a calculation path in combination with real-time data, and the resource distribution module redistributes resources when the calculation power exceeds the limit; the model compression module generates a compression model adaptive to the edge device through pruning and quantization algorithms; the parameter fine tuning module is used for jointly optimizing model parameters for a multi-task scene; according to the method, the adaptability and efficiency bottleneck of an existing debugging tool in a complex scene is effectively solved, and the intelligent debugging level of an automatic system of a hydraulic power plant is improved.

The invention relates to the technical field of model compression, and provides a large model compression method and device, a task processing method and equipment and a storage medium, and the large model compression method comprises the steps: carrying out the layer-by-layer quantification of a linear layer of a to-be-compressed initial large model, and obtaining a first large model; the initial large model is a pre-trained large language model constructed based on an expert hybrid architecture; performing route calibration on each expert sub-model in the first large model to obtain a second large model; in the reasoning process of the second large model, based on the task type of a to-be-executed target task, the importance of each expert sub-model in the task type is evaluated; and performing dynamic pruning on each expert sub-model based on importance so as to compress the second big model. Through a compression mode of combining static quantification and dynamic pruning, on the basis of ensuring the model performance, the memory and calculation overhead required by large model reasoning can be reduced, and efficient operation of the large model on light-weight equipment with limited video memory resources is facilitated.

The invention discloses a PDF (Portable Document Format) drawing recognition and information structured extraction method. The method comprises the following steps: generating a high-resolution bitmap through image preprocessing; positioning and classifying a text region, a table region and a symbol region in the drawing based on a target detection model of transfer learning; hough transform is combined with SIFT feature matching to identify engineering symbols, and sub-pixel positioning is realized through an RANSAC algorithm; after the oblique text is corrected through affine transformation, the content is extracted through OCR; reconstructing a table structure based on OPTICS clustering and projection analysis; constructing an RDF knowledge graph according to a coordinate association rule; and using U-Net difference to detect and position an omission area and complementing the omission area. According to the method, deep learning and image processing technologies are fused, the problems of low rotating text recognition rate, table structure loss and semantic association deficiency in a traditional method are solved, through lightweight model compression and TensorRT acceleration, the analysis accuracy is remarkably superior to that of the traditional method, and the method can be widely applied to the fields of constructional engineering, petrochemical engineering and the like and has wide application prospects. And the drawing information processing efficiency and the data integrity are improved.

The invention discloses a deep learning-based cold resistance evaluation system for medicago sativa L., and the system comprises a data generation module which is used for generating a phenotypic image and corresponding physiological data of medicago sativa L. under low-temperature stress through a diffusion model embedded with plant low-temperature response physical constraints; the evaluation model module is used for extracting cold resistance characteristics from the image and physiological data by adopting a causal-driven dual-channel adaptive network; and the decision module comprises a hierarchical model distillation unit and a federal reinforcement learning unit, and the hierarchical model distillation unit and the federal reinforcement learning unit realize joint training of model compression and decision strategy optimization through an edge-cloud collaborative architecture, output a cold-resistant decision and realize visualization through an augmented reality interface. The method can effectively solve the core problems of traditional medicago sativa cold resistance assessment in the aspects of data generation, model generalization, decision-making efficiency and the like.

The invention relates to the technical field of water level monitoring, in particular to a visual monitoring system for the condensate water level of a micro-hyperbaric oxygen chamber, which adopts the technical scheme that an internal image of a water receiving box is acquired at a preset frequency, and water mist and reflective interference are eliminated by adopting a dynamic illumination compensation and frequency domain filtering technology; a U-Net deep learning model is integrated, a water level line is segmented through multi-scale feature fusion, foam and impurity artifacts are recognized in combination with time sequence analysis, a water level value is dynamically corrected, graded alarm is triggered according to a water level threshold value, an air conditioner drainage system is linked, traditional image processing and the deep learning model are combined, and detection robustness under complex working conditions is improved. Real water levels and interferents are effectively distinguished through time sequence analysis and physical characteristic modeling, the real-time requirement under the high-pressure environment is met through the model compression and hardware acceleration technology, and absolute reliability of equipment operation is ensured through linkage of a grading early warning mechanism and a drainage system.

The invention discloses a remote sensing image classification model compression method based on incremental information guidance, and mainly solves the problems of evaluation deviation caused by insufficient representation of dynamic characteristics of a filter and poor coordination of pruning and quantization technologies in the existing model compression technology. Comprising the following steps: 1) acquiring a remote sensing image data set and preprocessing the remote sensing image data set; 2) pre-training the model by using the preprocessed data; 3) obtaining rank information and gradient information of the feature map, and calculating an importance score of a filter in the model according to the information; 4) optimizing affine transformation parameters, obtaining scores and generating a sorting result; and 5) performing compression and fine adjustment on the model in stages to obtain a lightweight model. According to the method, model compression is guided by gradient information, collaborative optimization is realized through integrated design of pruning and quantification in combination with rank information, the model scale and the calculation complexity can be remarkably reduced under the condition that the precision is not affected, and the compression effect is effectively improved.

The invention belongs to the technical field of computer graphics and real-time 3D scene reconstruction, and relates to a real-time compression and optimization method based on a 3D Gaussian sputtering (3DGS) model, which realizes efficient storage, transmission and rendering of a three-dimensional scene through residual vector quantization (RVQ) and a dynamic pruning technology, and comprises the following steps: S1, residual quantization vector quantization; s2, enhancing a dynamic pruning strategy; and S3, index compression and coding optimization. The invention provides a real-time 3D Gaussian sputtering (3D Gaussian sputtering) model compression system combining residual vector quantization and dynamic pruning, which can realize high-quality real-time rendering and extremely high storage efficiency. Through three core technologies of a vector quantization strategy, dynamic pruning strategy enhancement and index compression and coding optimization, on the premise that the rendering quality is guaranteed, the purposes that the model storage amount is reduced by more than 45 times, and the rendering speed is increased by 3 times are achieved.

The invention relates to a three-dimensional model compression transmission method and system based on dynamic feature perception, and relates to the technical field of three-dimensional live-action modeling, and the method comprises the steps: obtaining the original three-dimensional data of a three-dimensional model, carrying out the multi-modal feature extraction based on a model scene, dividing a key region, a transition region and a non-key region, and generating a three-dimensional model partition map, compressing each region according to the partition map label and a preset compression algorithm to generate a multi-resolution LOD sequence, determining a transmission data hierarchy in combination with a network state and equipment parameters, and finally rendering the model according to the transmission data hierarchy, user behavior information and an environment state to obtain a target three-dimensional model. The technical effects of effectively compressing, transmitting and rendering the three-dimensional model according to various factors such as different area characteristics, network states, equipment parameters and user behaviors of the model and improving the transmission efficiency and the rendering quality are achieved.

The invention belongs to the technical field of neural network architecture, and particularly relates to a dynamic reasoning path optimization method based on neural architecture search, which comprises the following specific steps: S1, designing a neural architecture search algorithm: firstly defining a search space, then selecting a search strategy, and then setting constraint conditions; s2, constructing a dynamic reasoning path: firstly performing input data feature analysis, then performing reasoning path guidance based on a knowledge graph, and then predicting the reasoning path by using a model according to the input data features and the knowledge graph; and S3, model training and optimization: firstly carrying out joint training, and then carrying out model compression and acceleration. According to the method, through neural architecture search algorithm design and dynamic reasoning path construction, the problem of computing resource waste is effectively solved.

The invention discloses a retraining-free pruning and recombination method for a sparse expert hybrid large model, and belongs to the technical field of large model compression and optimization. The method aims at solving the problems that due to the fact that an existing sparse expert hybrid (SMoE) model needs to load all expert parameters, memory occupation is too high, and deployment is difficult. According to the method, firstly, redundant experts are identified and pruned based on routing activation statistics; then, decomposing the pruned experts into neuron-level functional fragments, and redistributing the fragments to the reserved experts according to structural similarity; and finally, original fragments and newly distributed fragments are merged in the reserved experts through a weighted clustering algorithm, so that compact experts with fewer parameters and stronger expression ability are reconstructed. According to the method, fine-grained operation is carried out at the neuron level, the inherent representation conflict and dislocation problems among experts are effectively solved, the performance of the compressed model is remarkably improved, and reliable technical support is provided for deploying a large-scale SMoE model.

The invention discloses an air federated learning realization method for collaborative optimization of client scheduling and model compression. The invention realizes the air federated learning realization method for collaborative optimization of client scheduling and model compression. Along with rapid development of federated learning in a wireless network, air computing based on a multiple-input-multiple-output technology is widely concerned due to high communication efficiency of the air computing. However, in a resource-limited large-scale device scene, channel interference, device heterogeneity and limited spectrum resources significantly restrict the convergence rate and model performance of federated learning. In order to deal with the challenges, the invention provides an air federated learning framework combining compressed sensing and client scheduling, and by collaborative design of model parameter compression, multi-antenna beam forming and dynamic equipment selection strategies, the total communication overhead of each round of training is minimized, and meanwhile, the convergence of a global model is guaranteed. In order to balance training cost and model quality, a joint optimization problem based on calculation-communication cost and model precision loss is provided. In order to solve the non-convex optimization problem, an original problem is decoupled into two sub-problems. Firstly, an optimization problem of a pre-coding and post-processing matrix is designed for a given user scheduling result to minimize a gradient aggregation error. Then, a novel user scheduling algorithm based on channels and data is provided to obtain an air aggregation result.

The invention relates to the technical field of chip scheduling and optimization, in particular to a neural network-based main control chip task scheduling and dynamic performance optimization method, which comprises the steps of data acquisition and feature engineering, neural network model design, simulation environment training, model compression and deployment preparation, real-time state monitoring, dynamic decision reasoning, scheduling strategy execution and performance optimization. The data acquisition and feature engineering comprises the following steps: S1, hardware index acquisition; analyzing task attributes (calculation-intensive / IO-intensive), a dependency relationship (DAG), deadline (Deadline) and a resource demand (CPU / GPU occupancy rate); collecting data during chip operation through a performance counter (IPC, cache hit rate and branch prediction error rate), a temperature sensor and a power consumption monitoring unit (PMU); the neural network scheduler can achieve the energy efficiency ratio which is 20%-40% higher than that of a traditional method (such as a CFS scheduler), meanwhile, the neural network scheduler adapts to sudden load changes, and the practicability and the application range of a main control chip are wider.

A model compression method includes obtaining a first weight of each layer of a neural network model, where the first weight of each layer is a value of a floating-point type; and quantizing the first weight of each layer based on a quantization parameter to obtain a second weight of each layer. The second weight of each layer is a multi-bit integer. Quantities of bits of second weights of at least a part of layers are different. A quantity of quantization bits of a second weight of a layer with high sensitivity to a quantization error is greater than a quantity of quantization bits of a second weight of a layer with low sensitivity to the quantization error.

The invention relates to the technical field of large model algorithms, in particular to a large model dynamic compression optimization method and system based on sparse pruning, and the method comprises the steps: capturing original weight fluctuation data generated by resource fluctuation in reasoning, and obtaining sparse weight reference data through sparse processing; analyzing calculation complexity through model reasoning delay data, and separating reasoning delay amount caused by a model scale; dynamically controlling the model compression ratio within a preset performance range based on the delay amount and the sparse reference data, and collecting reasoning precision distribution data under different compression parameters; evaluating a model performance state under each parameter by means of a neural network simulation method, and generating a performance state simulation result; determining a model quality optimization compensation parameter based on a simulation result by combining resource fluctuation data acquired in real time in a compression process; and finally, the compression strategy is adaptively regulated and controlled through the compensation parameters, and collaborative optimization of model calculation complexity, reasoning precision and delay during dynamic change of hardware resources is realized.

The invention provides an equipment abnormal voiceprint detection system. The equipment abnormal voiceprint detection system comprises a voiceprint collection module used for collecting original voiceprint signals in real time and preprocessing the original voiceprint signals; the edge end detection module is deployed on edge computing equipment and is used for carrying out real-time anomaly detection on the voiceprint features by utilizing a one-dimensional lightweight neural network model; the abnormity credibility evaluation module is used for converting a real-time abnormity detection result into a probabilistic abnormity credibility score; the incremental data screening module is used for screening high-value samples from the real-time voiceprint data based on a dynamic density trend sensing algorithm and caching the high-value samples in the edge; the cloud model evolution module is used for carrying out evolution training on the detection model by utilizing a continuous learning mechanism; the generation and playback module is used for jointly generating pseudo samples through a variational auto-encoder VAE and a generative adversarial network GAN; and the model updating module is used for compressing the evolved cloud model and then issuing and replacing the original model in the edge end detection module so as to form a cloud-edge collaborative sustainable evolution closed loop.

The invention discloses a self-adaptive scene perception small target detection system, and the system comprises a scene perception and analysis module which is used for extracting scene features and calculating scene complexity; the multi-level feature extraction and reconstruction module is used for enhancing discriminative features in the scene features; the adaptive modal fusion module is used for dynamically generating a fusion weight for the enhanced scene features so as to perform feature fusion and iterative optimization; the spatio-temporal context modeling module is used for calculating a time sequence consistency constraint and a target appearance similarity, predicting a target motion track by using Kalman filtering, and establishing a spatial relation graph model; the detection strategy self-adaptive adjustment module is used for adjusting detection parameters of the spatial relation graph model so as to select an optimal detection strategy combination; and the edge calculation optimization module is used for constructing a knowledge distillation compression model so as to compress the spatial relation graph model. The method is high in environmental adaptability, can automatically adapt to different conditions, stably works around the clock, and is high in small target detection capability.

The invention relates to a model compression and environment perception method and device, equipment, a medium and a product. The method comprises the steps of obtaining a task scene of a target environment perception task and a to-be-compressed task processing model; the task processing model comprises at least one network layer, and the network layer comprises at least one channel; based on the task scene, determining an execution performance condition of the target environment sensing task; based on the execution performance condition, performing compression processing on the task processing model through a pruning strategy to obtain a target model; the compression processing is used for zeroing channel parameters corresponding to a to-be-pruned channel in the task processing model, the to-be-pruned channel is determined based on a pruning strategy, and the target model is used for executing the target environment sensing task and meeting execution performance conditions.