310 results about "Material classification" patented technology

The invention relates to the technical field of machine vision detection, and discloses a mobile phone shell production quality detection method and device based on machine vision. The method comprises the following steps: carrying out original multi-physical field data acquisition and space-time registration preprocessing on a mobile phone shell to be detected to obtain multi-modal standardized data; then, through surface texture region segmentation and edge contour extraction, correlation mapping of material classification and defect types is achieved, a mobile phone shell multi-modal feature vector is obtained, functional region differentiation feature fusion and injection molding process correlation reconstruction are conducted on the multi-modal feature vector, lightweight feature representation is obtained, and the mobile phone shell defect detection method is achieved. Generating an initial defect detection result through multi-class defect identification and quality grade classification; and performing defect visual feature extraction and feature comparison based on a historical qualified sample feature library to obtain defect deviation features, and generating a final quality detection result of the mobile phone shell to be detected through visual quality grade mapping. The accuracy and the process adaptability of mobile phone shell production quality detection are improved.

The invention relates to the technical field of computer vision, in particular to a station platform limit intelligent detection method and system based on computer vision, and the method comprises the following steps: obtaining platform region spectral data, extracting spectral features according to the reflectivity value of each pixel point in a spectral range, comparing the spectral features with the features of various known materials, and determining the position of the station platform limit. And identifying material types of the plurality of positions, detecting obstacles, and generating material classification feature values. According to the method, the spectral data is acquired, the reflectivity value is compared to identify the material type, the texture feature and the gradient change rate are utilized to analyze and identify the fuzzy area, misjudgment on the fuzzy identification is reduced, point cloud data and image information are adopted for time-space synchronous registration, and multi-source sensing information is fused to enhance the time-space consistency of three-dimensional reconstruction. The size accuracy of the platform model is improved, the space offset is calculated by comparing engineering design drawing data, a structure deviation quantitative evaluation system is established, a deformation area is positioned, and the foreseeability of safety early warning is improved.

When x-ray fluorescence (“XRF”) spectroscopy is utilized to classify materials transported on a moving conveyor belt, there is the possibility of the x-ray beam only partially irradiating the material piece, which can result in the capture of an inaccurate XRF spectrum needed to classify the material piece. This can lead to an improper (erroneous) classification and resultant sortation of material pieces (e.g., aluminum alloys). In a material handling system, the area of the intersections between the x-ray beam spots from an x-ray fluorescence system and the material pieces are measured and correspondingly used to correct the measured XRF spectrum associated with each material piece. The material pieces can then be sorted according to the corrected XRF spectrum.

The invention relates to a railway material purchasing asynchronous task processing method, device and system, and the method comprises the steps: constructing a six-layer message channel architecture of a purchasing business process, and executing an instruction of deploying a message agent node in an active-active mode for each channel based on the core message middleware of a Kafka cluster; the front-end gateway is instructed to package the purchase event into a binary message body with a transaction ID and a timestamp, and the binary message body is written into a persistent queue after priority labeling is carried out according to a task type label; processing the concurrent tasks by adopting a multi-stage feedback queue scheduling algorithm, and dynamically allocating computing resources based on a thread pool; a maximum retry threshold value is set according to an exponential backoff retry strategy, so that the failed task is automatically transferred into a dead message queue; and dynamically calculating an optimal message routing path based on the supplier portrait and the material classification model, and extracting the order to an exclusive processing queue of the optimal supplier according to Topic dynamic switching of the message queue. According to the invention, the coupling degree is reduced, the system performance is improved, and the fault tolerance is enhanced.

The invention provides an imaging method and system based on a polarization imaging lens and visual collaborative optimization, and a medium, and the method comprises the steps: integrating a four-way polarization filter array on the surface of an imaging sensor, and synchronously collecting a four-channel polarized light intensity image through single-frame exposure based on the imaging sensor; obtaining a three-channel RGB image, and inputting the four-channel polarized light intensity image and the three-channel RGB image into a polarization-RGB fusion neural network; polarization features and RGB features are extracted, heterogeneous information coupling optimization is carried out, and fusion features are obtained; constructing a polarization-color joint decoding network, carrying out polarization and RGB information collaborative decoding on the fusion features, training a polarization decoding process and a material classification task together based on an end-to-end joint optimization strategy, and outputting a material classification result; high-efficiency acquisition of original polarization information on a hardware level is guaranteed through the polarization filtering array; and an end-to-end joint training strategy is adopted to realize optimal decoupling and fusion of perception information.

The invention provides a universal method for automatically identifying woven fabric density based on deep learning, and relates to the field of artificial intelligence and image analysis and processing. The deep learning-based universal method for automatically identifying the density of the woven fabric comprises the following steps of: a, acquiring a high-resolution color woven fabric image by an image preprocessing module, and carrying out graying processing on the high-resolution color woven fabric image; the method comprises the following steps: automatically identifying the inclination angle of the yarn and carrying out coordinate point transformation to obtain an inclination-corrected grayscale image, obtaining a grayscale projection discrete curve in the longitude and latitude directions of the image and carrying out smoothing processing by adopting Gaussian filtering to obtain a minimum value point of discrete data; the universal method for automatically identifying the woven fabric density based on deep learning has the advantages that 1, the automation degree is high: the end-to-end density automatic identification is realized based on the deep learning technology, the manual intervention is reduced, and the detection efficiency and accuracy are improved; compared with a traditional method, the steps of manual proofreading and material classification are avoided, and the whole recognition process is more efficient.

The invention relates to a terahertz spectrum-based substance classification method and device. The method comprises the following steps: providing a reference substance group, constructing a terahertz spectrum classification model for substance classification based on the reference substance group, providing a to-be-classified target substance, and performing terahertz spectrum processing on the target substance to generate an absorption spectrum characteristic pattern of the target substance based on the terahertz spectrum processing on the target substance; and loading the absorption spectrum characteristic pattern of the target substance into the constructed terahertz spectrum classification model so as to perform identification and classification processing on the absorption spectrum characteristic pattern of the target substance by using the terahertz spectrum classification model, and generating the category of the target substance after identification and classification processing. Based on the terahertz spectrum, the requirement for material classification can be effectively met, the classification precision is improved, and the consumption of hardware resources is reduced.

The present invention discloses a method for material classification and attribute extraction in the construction industry based on a large language model, including constructing a traditional database and a vector database based on a standard classification system data set and a full category data set; inputting irregular text, and retrieving the irregular text based on a large language model using a standard classification name vector database, a standard classification sample vector database, a full category classification name vector database, and a full category classification sample vector database to obtain retrieval information; performing preliminary matching and secondary matching on the retrieval information based on the large language model to obtain a material classification name; and extracting attributes of the material classification name from a traditional database based on the large language model to obtain material attributes. The present invention utilizes the semantic understanding ability of the large language model and the information retrieval mechanism to integrate multi-source data, thereby achieving efficient and accurate material classification and attribute extraction.

The invention provides a bottle sorting and data monitoring method and system and bottle recycling equipment. The method comprises the following steps: collecting a visible light image and near infrared spectrum data of a bottle target, and preprocessing the collected visible light image and near infrared spectrum data; extracting a fusion spectrum feature vector of the pre-processed visible light image and the near infrared spectrum data, and generating a feature matrix for material classification; inputting the feature matrix into a lightweight convolutional neural network model deployed in an edge calculation unit, and obtaining a recognition result of the material and form of the bottle; a sorting control instruction is generated based on the recognition result, and motion space state representation is constructed in combination with current state information of the mechanical arm; and inputting the spatial state representation into a path optimization model based on deep reinforcement learning, and outputting and executing an optimal sorting path and a coordination action sequence. According to the method, the precision and robustness of bottle identification can be improved, the recycling efficiency of renewable resources can be improved, and the recycling cost can be reduced.

The invention relates to the technical field of artificial intelligence and deep learning, in particular to a material intelligent classification and deletion detection system based on a deep learning model. In an existing insurance claim settlement process, various kinds of materials need to be submitted, and checking and auditing are mostly completed manually. At present, an insurance company starts to use a deep learning technology to assist auditing, the model scale is huge, a large storage space and strong GPU computing power support are needed, and the cost is high. The system is composed of a material preprocessing module, a hierarchical classification recognition module, a deletion detection module and a result output and prompt module, and all the modules interact through standardized data interfaces; the hierarchical classification and recognition module is used for pre-classifying uploaded materials by using a hierarchical cascade classification architecture and a lightweight convolutional neural network, and quickly pre-classifying the uploaded materials according to scene-class and character-class pictures; the lightweight intelligent classification and deletion detection system provided by the invention gives consideration to high precision and high efficiency, meets actual service requirements, and cannot be replaced by a traditional scheme.

The invention relates to the technical field of data processing, and discloses a professional warehouse control method and system based on artificial intelligence and a storage medium. The method comprises the steps of collecting professional warehouse data through a multi-source sensing device, and performing fusion to obtain a physical resource multi-source fusion data set; carrying out classification by adopting a transfer learning algorithm to obtain engineering material classification result data; mapping the reserve quota standard to obtain state evaluation data; carrying out demand prediction to obtain deployment scheme data; decomposing the operation process to obtain operation instruction data; and performing carbon emission analysis to obtain environment management decision data. According to the invention, the efficiency and accuracy of professional warehouse control based on artificial intelligence are improved.

The invention discloses a building industry material classification and attribute extraction method based on a large language model. The method comprises the following steps: constructing a traditional database and a vector database based on a standard classification system data set and a full-category data set; inputting an irregular text, and retrieving the irregular text by adopting the standard classification name vector database, the standard classification sample vector database, the full-category classification name vector database and the full-category classification sample vector database based on a large language model to obtain retrieval information; performing preliminary matching and secondary matching on the retrieval information based on a large language model to obtain material classification names; and carrying out attribute extraction on the material classification names from the traditional database based on a large language model to obtain material attributes. According to the method, multi-source data are integrated by utilizing the semantic comprehension ability of a large language model and an information retrieval mechanism, and efficient and accurate material classification and attribute extraction are realized.

The invention belongs to the technical field of data management. According to the material data cleaning method and system fusing the large language model and the knowledge graph, the knowledge graph stores core knowledge such as an enterprise material classification system, term definition and quality rules in a structured triple form, and a clear standard basis is provided for cleaning operation; accurate structured knowledge anchor points and standardized constraints are provided for the large language model, the illusion of the large language model is effectively inhibited, and it is ensured that the output content conforms to industry specifications; the big language model injects strong natural language understanding, fuzzy semantic analysis and unstructured information extraction capabilities into the knowledge graph, makes up for the flexibility deficiency of the knowledge graph, and deeply fuses the two to jointly solve the key problem of what the text semantics is understood and whether the semantics meets the standard or not. And the material data cleaning precision is improved.

The invention provides a multi-layer cascade express package image identification method, which comprises the steps of collecting express package image data, constructing a multi-label express package data set, pre-training a YOLOv8 framework model, migrating the pre-trained model to a corresponding sub-data set in the multi-label express package data set, and identifying the express package image according to the sub-data set. Respectively constructing an internal and external package classification model, an internal cushion package classification model, an external package material classification model, an external packaging mode classification model and a multi-target positioning identification model, judging the type of the to-be-identified image, if the to-be-identified image is a single-target package image, inputting the to-be-identified image into the internal and external package classification model, and if the to-be-identified image is a single-target package image, outputting the to-be-identified image; and inputting an output result of the internal and external package classification model into an external package material classification model, an external packaging mode classification model and an internal cushion package classification model for identification, and if the image is a multi-target package image, inputting a to-be-identified image into a multi-target positioning identification model for identification of the multi-target package image. According to the invention, express package image identification can be realized.

The utility model discloses quartz sand filter material grading and screening equipment, which comprises a box body, a first screening plate, a second screening plate, a first screening plate and a second screening plate, and the inclined baffle is obliquely arranged between the first screening plate and the second screening plate to bear quartz sand falling from the first screening plate, and the low-position end of the inclined baffle faces the high-position end of the second screening plate. According to the quartz sand filter material grading and screening equipment provided by the utility model, quartz sand particles screened from the first screening plate can be prevented from directly impacting the second screening plate and directly popping out of the second discharging hole by utilizing the existence of the inclined baffle, so that the screening quality of quartz sand is improved; and secondly, the inclined baffle can guide the falling quartz sand particles to roll towards the high-position end of the second screening plate, so that the horizontal moving distance of the quartz sand particles on the second screening plate is increased, and the screening quality of the small-particle quartz sand is improved.

The invention discloses a collaborative optimization method and system for classified disposal and green disassembly of power grid waste and old materials, and relates to the technical field of classified disassembly of materials, and the method comprises the following steps: obtaining multi-source feature data of to-be-disposed power grid waste and old materials; constructing a residual value prediction model based on the multi-source feature data, and outputting a component residual value mapping graph; analyzing the structure information of the target material, and generating a disassembly path decoupling graph in combination with the residual value mapping graph; establishing a dual-objective optimization scheduling model, and solving the scheduling model to generate a green disassembly operation sequence meeting a first constraint condition; and when the disassembling operation sequence is executed, collecting real-time feedback data, dynamically correcting residual value prediction model parameters and scheduling model constraint weights. According to the method, the disassembly path decoupling graph and the disassembly state transition graph are constructed, collaborative representation of the component value and the disassembly topology in the equipment is realized, and the problems of decoupling of residual value evaluation and path planning and lack of green evaluation in a traditional method are effectively solved.

The invention relates to a simple MRP system and method based on a report platform, which can solve the problems of data isolation, inflexible inventory management and the like of the traditional MRP system. The system integrates multi-source data by using an ETL tool, can dynamically calculate safe stock, sets parameters according to material classification, and triggers replenishment in combination with stock and arrival conditions. The conflict pre-judgment module can model supplier delivery time and generate a multi-scheme comparison report. Real-time neat set rate monitoring can analyze according to materials / processes, and production interruption can be prevented by using a work order state machine. And the rules can be continuously optimized through substitute optimization, inventory coefficient self-calibration and the like.

The invention relates to the technical field of power equipment state monitoring and fault diagnosis, and discloses an infrared diagnosis method and system for power transmission and transformation equipment. The method comprises the following steps: acquiring infrared thermal image data of equipment and removing noise by using a filter to obtain a clear image; carrying out pixel-level segmentation by adopting an image segmentation model, determining the boundary of the equipment part, and extracting an initial feature vector which comprises texture and depth heat distribution information; based on the feature sets, judging a temperature abnormal region, and performing material classification in combination with texture information; according to the method, the structural difference of the components is analyzed through the classification model, structural feature description is obtained, the structural features are further combined with the temperature abnormal area, fine feature matching is carried out, and therefore synchronous recognition is achieved, and the accuracy and real-time performance of power transmission and transformation equipment fault diagnosis are improved.

The invention discloses an intelligent examination and approval management method for the housing and construction industry, and the method comprises the steps: collecting examination and approval application data, carrying out the data cleaning, data deduplication and data standardization operation of the examination and approval application data, and obtaining preprocessed data; constructing a data classification model based on a machine learning algorithm, and performing material classification on the preprocessed data by using the data classification model to obtain a material classification result; according to a material classification result, performing primary processing and risk assessment on the preprocessed data by using an intelligent auditing system to obtain a risk score; performing secondary processing on the pre-processed data by using an expert system to obtain a personalized approval scheme; and examining and approving the preprocessed data according to the personalized examination and approval scheme, and storing the risk score and the examination and approval result to an intelligent examination and approval system. The invention relates to the technical field of examination and approval management, and solves the technical problems of low data quality, lack of flexibility and intelligence and the like of an existing residence construction examination and approval system.

The invention discloses a material classification method and device based on an ultraviolet electron spectrum, a medium and a product, and relates to the technical field of material surface science and data analysis. The method comprises the following steps: acquiring ultraviolet light electron energy spectrum data of a target material to be analyzed, and extracting a first spectrum curve from the energy spectrum data; obtaining a conductor characteristic identification result of the target material according to the first spectrum curve; under the condition of determining that the target material does not have the conductor characteristics based on the conductor characteristic identification result, extracting a second spectrum curve in a preset valence band top energy interval extending towards the negative energy direction by taking the Fermi level as a starting point from the ultraviolet light electron spectrum data, and performing multi-dimensional characteristic fusion on the second spectrum curve, generating a semiconductor characteristic identification result of the target material based on the fusion result; and determining a classification result of the target material according to the semiconductor characteristic identification result. According to the scheme, the conductive attributes of the target material can be quickly and accurately classified.

Provided are systems and methods that perform material classification of imagery using generative denoising diffusion models. Traditional material classification systems, which are predominantly based on discriminative models, face issues with generalization and diversity, particularly when encountering new or complex material compositions. Furthermore, these systems often function like black boxes, making it difficult to understand their decision-making processes or identify spurious correlations. The provided systems and methods address these challenges by leveraging the capabilities of diffusion models, which can provide better generalization and more transparent decision-making processes.

The invention discloses an intelligent insurance claim settlement processing method and device based on AI, equipment and a medium. The method comprises the following steps: constructing an insurance product digital model and a claim knowledge base; deploying a multi-terminal intelligent claim settlement channel, and receiving a claim settlement application and a multi-modal claim settlement material submitted by a user through the claim settlement channel; an intelligent material processing engine is adopted to carry out analysis processing on the multi-modal claim settlement material to obtain structured claim settlement data, and analysis processing comprises material classification, image optimization, key information identification and medical text semantic analysis; based on the insurance product digital model, the claim settlement knowledge base and the structured claim settlement data, claim settlement auditing is executed through an AI intelligent auditing decision flow, and a claim settlement auditing result is generated; and the claim settlement audit result and the claim settlement progress information are fed back to the user through the multi-terminal intelligent claim settlement channel, and the claim settlement operation is executed, so that the problems of difficulty in material identification, dependence on manpower in the process, low efficiency and the like in traditional insurance claim settlement can be solved, and intelligent processing of the whole process of insurance claim settlement is realized.

The invention provides a split type material tray, packaging device detecting, transferring and classifying equipment and a detecting method, and belongs to the technical field of packaging device detection.The split type material tray comprises a mother tray and a son tray, edge steps are arranged on the periphery of the mother tray, and object taking notches facilitating taking out of the son tray are symmetrically formed in the periphery of the mother tray; the sub-trays are provided with positioning grooves distributed in an array mode, and the sub-trays are stacked in the mother tray. The packaging device detecting, transferring and classifying equipment comprises an equipment rack, a feeding tray, a discharging tray, a conveying belt, a detecting module, a positioning camera, a feeding robot, a discharging robot and a waste box. The tray is composed of the mother tray with the unified specification and the sub-trays with different sizes and different grades, when the specifications and the sizes of the detected packaging devices are different, only the matched sub-trays need to be replaced, and the whole tray does not need to be replaced; and the robot transfers the materials between the tray and the working area through the positioning camera and coordinate transformation, and the qualified materials and the unqualified materials are classified and stored.

The invention provides a steel product process-level process modeling carbon footprint evaluation system and an analysis method thereof, and relates to the technical field of carbon emission data analysis. The steel product process-level flow carbon footprint evaluation system comprises a material management module, an emission factor management module, a process module and a carbon footprint calculation module. The material management module is used for configuring material codes, material classification, transportation types and material consumption; the emission factor management module is used for managing attributes of production process emission factors and hidden emission factors of materials; and the process module is used for adding input and output related information and setting emission factor related information. By adopting process-level process modeling, the problem of coarse granularity of a carbon footprint accounting result in a traditional black box calculation mode is effectively solved, the calculation process is transparent through process-level process modeling, and the life cycle carbon footprint value of each intermediate product in the steel production process can be traced.

The invention particularly relates to an intelligent garbage classifying and sorting system based on machine vision, and relates to the technical field of intelligent garbage classification and industrial automation. A digital twin construction module; an adaptive control and optimization module; and a predictive maintenance and health management module. According to the method, vision-touch-acoustics multi-mode perception is deeply fused, so that the identification limitation of a traditional single vision sorting system is solved; after the visual identification and the tactile feature are subjected to weighted fusion, the material classification accuracy is improved, and the problems of misjudgment and missed judgment under the scenes of oil contamination, light reflection, form deformation and the like are effectively avoided.

The invention discloses an application material intelligent classification method and system based on data analysis. The method comprises the steps that after an application material text is obtained, keywords are extracted, weights are calculated, and weighted keyword vectors are generated; and mapping the vector to a high-dimensional semantic space, and converting the vector into a multi-dimensional semantic representation vector. And calculating a semantic similarity sequence of the vector and a preset category standard vector, and performing preliminary classification by comparing the semantic similarity sequence with a high similarity threshold. And if the matching is successful, outputting a corresponding type as a final result. If the keyword fails, the keyword weight is evaluated again, and the enhanced semantic representation vector is generated in an enhanced mode. And calculating the similarity between the abnormal material and the standard vector, and screening out the abnormal material according to a low threshold value. And finally, performing semantic proximity analysis and consistency evaluation grouping on the abnormal materials to obtain final classification. According to the method and the system, the classification accuracy of various application materials in the fields of digital public services and the like can be effectively improved.

The invention provides a building representation carbon footprint estimation method and device based on artificial intelligence recognition and speculation, and belongs to the technical field of carbon emission estimation. The method comprises the following steps: collecting building data to obtain a structure classification model data set and a material classification model data set; the method comprises the steps of calculating building feature values, establishing a building structure prediction model based on machine learning, replacing a feature extraction part of a YOLOV11 model with a SwinTransformerV2 or ConvNeXtV2 computer vision network based on the YOLOV11 model, and establishing a building material prediction model based on deep learning; collecting current situation data of a to-be-evaluated building, and obtaining a current situation characteristic value of the building through the model; constructing a carbon emission calculation model based on the building current situation characteristic values and the carbon emission factors; and calculating an estimated value of the building carbon emission. According to the method, the building material and carbon emission can be estimated more conveniently, efficiently and accurately.

The invention discloses an electronic material automatic classification method and system based on multistage feature recognition, and relates to the field of material classification, and the method comprises the steps: firstly obtaining unstructured material information, and carrying out the regularization and preprocessing of the information; subsequently, through main category identification and feature lexical element marking, accurate positioning of material core attributes is realized, which exceeds shallow identification of traditional keyword matching, and the defect that complex semantics cannot be processed is effectively overcome. On the basis, structured extraction and analysis are further carried out on the marked features, and the marked features are converted into paradigm features which can be deeply understood by a machine. And finally, through code integration and verification, a standard and unique production code, such as an SPYY. XBBBVVVV structure, is automatically generated. In this way, low-efficiency and error-prone manual operation is replaced, and the fundamental problem that an existing automatic scheme is difficult to deal with description diversity is solved through the powerful analysis capacity and robustness of the manual operation.

The utility model provides a material classification and recovery device, which relates to the technical field of cigarette preparation and comprises a material tank, an inspection part and a recovery part, and the inspection part comprises a metering tank, a weighing device and a liquid level meter; the recycling part comprises a first recycling pipe and a second recycling pipe, a recycling barrel is arranged at the lower end of the first recycling pipe, a rechecking barrel is arranged at the lower end of the second recycling pipe, a first valve is arranged at the conveying pipe, a second valve is arranged at the first recycling pipe, and a third valve is arranged at the second recycling pipe; the first valve, the second valve and the third valve are all electrically connected with the controller. The device is simple in structure, through cooperation of the three valves and weighing and liquid measuring at the metering tank, the material returning condition can be intelligently detected, automatic calculation and comparison of the material returning precision and the material liquid density are achieved, manual rechecking is informed when data are abnormal, reasonable and stable material returning is guaranteed, manpower consumption is reduced, and stable tobacco production is also guaranteed.

The utility model belongs to the technical field of production equipment, and discloses a material receiving, detecting and basketing system which comprises a material distributing unit, a temporary storage unit, a spot check unit, a material arranging unit and a basketing station. The sampling inspection unit comprises a detection platform, a detection module, a classification module, a qualified product platform and a waste collection platform, the material distribution unit selectively places materials on the cache unit or the detection platform, the detection module is used for detecting the materials, and the classification module is used for transferring the materials on the detection platform to the qualified product platform or the waste collection platform; the basket loading station is provided with a material basket, and the material placing unit can move among the temporary storage unit, the qualified product platform and the basket loading station and is used for picking up materials of the temporary storage unit and the qualified product platform and transferring the materials to the material basket. According to the system, the working procedures of detecting, classifying, basketing and the like of products can be rapidly achieved after the products are discharged, the collaboration among the mechanisms is improved, the basketing efficiency is improved, the layout can be more reasonable, and cost reduction and efficiency improvement are achieved.