37results about How to "Effective classification" patented technology

The utility model discloses a kind of alkali-etching rework piece storage devices for solar cell processing, including storage table, sliding plate, first limiting component and second limiting component, multiple storage grooves are provided on storage table, sliding plate is slidably connected in storage groove, first limiting component and second limiting component are all installed on sliding plate, first limiting component and second limiting component are vertically arranged, and placement space is formed between first limiting component and second limiting component, storage box is placed in placement space, one end of sliding plate is fixedly connected with the baffle matched with storage groove, the outer diameter of storage groove is less than the outer diameter of baffle, and label is detachably installed on baffle.Compared with prior art, a kind of alkali-etching rework piece storage devices for solar cell processing of the utility model, it can place multiple storage boxes, by placing different specifications solar cells in different storage boxes, effective classification and management are realized.

The invention discloses a computer data management system and method, and relates to the technical field of data management, and the method comprises the steps: reading a heterogeneous data set, dividing the heterogeneous data set into text, time sequence and binary data groups according to data types, extracting a core field of each data group, and generating a deduplication feature and an index feature, setting a sliding window based on the data type and generating a window index feature abstract; calculating a comprehensive similarity score between the data blocks according to the sliding window, screening a similar data block set, determining a reference data block and a redundant data block, clustering the reference data block to generate an initial cluster, and if the global deduplication rate and the clustering preprocessing validity do not meet preset conditions, re-initializing the sliding window; according to the method, the cluster structure is optimized and the cluster feature abstract is generated based on the retrieval popularity predicted value of the initial clustering cluster, the three-layer index structure comprising the global index, the intra-cluster index and the feature index is constructed, the retrieval efficiency is tested, the index hierarchy is optimized, and the storage efficiency and the retrieval performance of heterogeneous data management are remarkably improved.

The application discloses a data search and classification method based on eigenvalue index, comprising the following steps: obtaining a set based on original data through a regular semantic hybrid parser, wherein the set comprises feature names and feature values; obtaining a classification vector based on the set through mapping; obtaining classified index data based on the classification vector through two-level index search of multidimensional reverse arrangement and bitmap; obtaining a query data subset based on user query conditions through query rewriting, feature sorting and bitmap intersection; obtaining an updated query data subset based on the query data subset through sequential scanning and sorting; and obtaining updated classified index data based on the query data subset and new data through a bitmap merging strategy.

The application discloses a method for extracting keywords information of bidding and opening bidding based on regular expression and scrapy, comprising the following steps: S1, determining the analysis field of the bidding and opening bidding project; S2, collecting and sorting the keyword list of the field, wherein the keyword list is created as an ordered keyword list or an unordered keyword list according to the field requirement; and S3, selecting the subsequent processing method according to the specific field condition. By combining the regular expression and the scrapy, the application can extract multiple keywords information of the bidding and opening bidding project, so as to avoid the problem of missing search. The application respectively extracts the keywords information of the bidding, the winning bidding and the opening bidding project, can carry out targeted query extraction according to the different keyword extraction modes of the opening bidding project, realizes the quick extraction of important information in the text, removes the useless data, solves the problems of information redundancy, reading difficulty and the like.

The application provides a video classification method and device, an electronic device and a storage medium, wherein the method comprises: encoding a first video to obtain a plurality of corresponding image frames, wherein the first video is any video in a plurality of videos to be classified; determining the complexity of the first video based on a target image frame in the plurality of image frames, wherein the complexity comprises spatial complexity and temporal complexity; and classifying the plurality of videos based on the complexity. Through the application, the technical problem that the prior art classifies by using the bottom features of a video and needs to rely on strong prior knowledge, resulting in poor video classification effect and weak robustness, is solved.

The utility model discloses a PCB (Printed Circuit Board) classifying and transporting device and conveying equipment thereof, comprising a classifying and transporting module, the classifying and transporting module is used for moving a first PCB and a second PCB which are classified, and the classifying and transporting module is provided with a first table body, a first transporting channel and a second transporting channel; the first transportation channel and the second transportation channel are both arranged at the upper part of the first table body, and the first transportation channel is used for moving a first PCB to the second transportation channel; the PCB recovery module is provided with a first PCB transfer unit, an inlet of the PCB transfer unit is connected with the classified transportation module, and the first PCB transfer unit is used for storing and transferring second PCBs; and the PCB detection module is connected with the outlet of the first PCB transfer unit, and the PCB detection module is used for detecting and repairing the second PCB. According to the technical scheme of the embodiment, good products and defective products of the PCBs can be effectively classified.

This application discloses a method, apparatus, device, storage medium, and program product for classifying complaint work orders, aiming to solve the problems of low processing efficiency, insufficient classification accuracy, and high labor costs caused by the reliance on manual classification of complaint work orders. The method includes: acquiring user complaint text data; inputting the complaint text data into a trained work order classification model, and outputting the classification result of the complaint text data; wherein the work order classification model includes at least a feature extraction layer, an attention layer, and a classification layer; the feature extraction layer is used to extract text features at least two scales from the complaint text data, and fuse the text features at least two scales to obtain fused features; the attention layer is used to assign weights to features corresponding to different words in the fused features to obtain weighted features that reflect the differences in word importance in the complaint text data; the classification layer is used to output the classification result based on the weighted features.

The application belongs to the technical field of image recognition segmentation, and discloses a medical image cell segmentation and tracking method, which comprises the following steps: step 1: data processing; feature extraction: the backbone part in the model is used to extract features from the preprocessed image, and the CSPDarknet structure is adopted in YOLOv8; step 3: FPN-PAN multi-scale feature fusion; step 4: Head prediction according to multi-scale features. The application realizes real-time tracking of the motion trajectory of cells by combining with a tracking algorithm such as deepsort. The method is mainly based on the YOLOv8 framework, and the Simam attention mechanism and the multi-scale proto method are adopted to optimize the model, so that the detection effect of YOLOv8 is further improved. The application can automatically complete the analysis and detection of medical images, is high in convenience and easy to use.

The application provides a colorectal cancer staging algorithm and system based on a self-supervised learning mode, which can judge the severity of colorectal cancer without using labeled samples. First, image transformation and low-level processing are performed on the data set to generate different groups of new data sets. The different groups of data sets are input into different encoder networks to obtain feature outputs, and the image information is saved through the conversion of the conditional attention mechanism and the cross-model hybrid maximum. Finally, the model is trained and optimized by merging and calculating the contrast loss and the reconstruction loss. The application can effectively classify the severity of colorectal cancer without a labeled training set by learning the similarity and difference of medical images to construct features, and has a wide application prospect.

The present application relates to a kind of small sample disease classification methods based on association learning, comprising: first, build a small sample asphalt disease data set containing multiple asphalt disease types;Second, using the principle of association learning, develop a set of deep learning model and its training strategy suitable for small sample disease classification;Then, by setting "precision-speed" comprehensive evaluation index and a variety of damage simulation mode, evaluate the comprehensive performance and robustness of the model;Finally, the model is further adjusted and optimized, and deployed to road disease inspection equipment, realize the accurate classification of road disease.Compared with the prior art, the present application aims to solve the problem that traditional deep learning method excessively depends on training data size, alleviate the deficiency of existing method under small sample condition, improve the classification accuracy and system robustness, and provide a new solution for road disease detection and classification.

The building waste classifying and recycling device comprises a recycling box, a driving motor, crushing rollers, a primary screening drawer and a screening plate, the bottom of the recycling box is of an open structure and is fixedly provided with a plurality of supporting legs, the driving motor is fixed to the rear side wall of the recycling box, and the multiple crushing rollers are rotationally connected into the recycling box; roll shafts of the multiple crushing rollers penetrate out of the rear side wall of the recycling box and are in transmission connection with a power output shaft of the driving motor. The primary screening drawer is slidably connected into the recycling box in a pull-out mode and can be pulled out from the front side of the recycling box, a plurality of first screening holes are formed in the primary screening drawer, and a plurality of magnetic blocks are fixed to the bottom of the primary screening drawer; the sieve plate is slidably connected into the recycling box and horizontally arranged, the sieve plate is located below the primary screening drawer, a plurality of second sieve holes are formed in the sieve plate, and the hole diameter of the second sieve holes is smaller than that of the first sieve holes; the two ends of the sieve plate stretch out of the left and right side walls of the recycling box by a preset distance. One end of the sieve plate is in transmission connection with a power output shaft of the driving motor through a reciprocating mechanism.

The utility model provides powder recovery equipment for metal 3D printing, which comprises a recovery box provided with a feed inlet for metal powder to enter the recovery box, and a partition plate for dividing the internal space of the recovery box into two recovery cavities, the recycling box is provided with two discharging ports corresponding to the two recycling cavities and an air outlet channel used for air outlet. And the fan is arranged on the air outlet channel and used for leading airflow into the air outlet channel from the feeding port and enabling the metal powder entering the recycling box to enter different recycling cavities. Metal powder enters different recovery cavities according to different particle masses through airflow, the problem that a traditional screening plate is blocked due to the fact that large-particle powder is gradually increased is solved, therefore, powder can be classified and recovered more efficiently, and the overall powder recovery efficiency is improved.

The utility model discloses a detector wire harness comprising a connector, the connector is connected with a plurality of first lines, one end of each first line far away from the connector is branched to form a plurality of second lines, and one end of each second line far away from the first line is respectively connected with a pin terminal. The detector wire harness further comprises a plurality of first sleeves and rubber sleeves, the first sleeves respectively wrap the outer walls of the first circuits, and the rubber sleeves are distributed at intervals and used for tightly sleeving the ends, close to the connector, of the first circuits into a bundle. According to the utility model, only one connector is needed to be connected with different components, the use is convenient, disorder or mess is not easily caused when cables are connected, the wiring is reasonable, the wiring arrangement is orderly, the wire harness integration degree is high, and the maintenance is also convenient.

The utility model relates to the technical field of alumina washing, specifically is a kind of alumina multistage filtration washing device, including machine body, the left side middle part of machine body is fixedly connected with feed seat, the right side of machine body is fixedly connected with discharge seat, the inner wall of machine body is movably connected with rotating stand, the outer wall inside of rotating stand is fixedly connected with filter plate, the left side upper of machine body is fixedly connected with transmission case, the left side front of machine body is fixedly connected with control cabinet, the inside of transmission case is equipped with rotating mechanism, the inside upper of machine body is equipped with flushing mechanism. The utility model has realized the flushing of raw material rotation filtering and filter plate, has promoted the high efficiency and convenience of device.

The invention discloses an exhibition subject identifier classification method and system fusing partition color features, belongs to the technical field of image recognition, and can solve the problem of low classification precision in the prior art. The method comprises the following steps: S1, extracting a first local area corresponding to each connecting piece and a second local area corresponding to each label from a monitoring image in an exhibition hall by using a target detection model; s2, determining a first effective pixel meeting a first preset condition according to the first local area, and determining a second effective pixel meeting a second preset condition according to the second local area; the first preset condition comprises saturation greater than or equal to preset saturation, and the second preset condition comprises saturation greater than or equal to preset saturation and brightness less than or equal to preset brightness; s3, determining a combined hue of each exhibition subject identifier according to the first effective pixel and the second effective pixel; and S4, classifying all the exhibition subject identifiers according to the combined hues. The method is used for counting the number of exhibitors.

The utility model discloses an efficient granulator for colored master batch production, relates to the technical field of colored master batches, and aims to solve the technical problem that substances adhered to a cutting knife cannot be cleaned in time. According to the technical scheme, the cutting device is characterized by comprising a rack, a feeding port is formed in one side of the rack, a discharging port is formed in the opposite side of the rack, a discharging guide plate is arranged on the discharging port, a plurality of flow guide grooves are formed in the discharging guide plate, the flow guide grooves are communicated with a waste port, a cutting cavity is formed in the rack, and a conveying mechanism is arranged in the cutting cavity; four cutting telescopic sleeves are arranged at the discharging end of the conveying mechanism, each cutting telescopic sleeve is elastically connected with a cutting telescopic rod through a telescopic spring, a cutting lifting mechanism is arranged on the four cutting telescopic rods, a cutter is arranged on the cutting lifting mechanism, and a cleaning mechanism is arranged on the cutting telescopic sleeve on the same side. And the conveying mechanism and the cutting lifting mechanism are driven by a driving mechanism. The utility model aims to provide an efficient granulator for producing colored master batches.

The invention discloses a seismic exploration data unsupervised similarity clustering method, electronic equipment, a storage medium and a device. The method comprises the following steps: constructing a seismic exploration data unsupervised similarity clustering model based on a self-organizing mapping network; constructing a training sample and a test sample based on actual seismic exploration data; training the clustering model based on the training sample; testing the trained clustering model based on a test sample; and carrying out seismic exploration data unsupervised similarity clustering based on the tested clustering model. According to the method, the seismic exploration data unsupervised similarity clustering model is constructed through the self-organizing mapping network to carry out seismic exploration data unsupervised similarity clustering, data are classified and identified according to geophysical rules and standards, effective and rapid classification of seismic data can be realized, the seismic data characterization capability is improved, and the seismic exploration data classification efficiency is improved. And a basis is provided for accurate processing and explanation of seismic exploration data.

The invention provides a light variation curve classification method based on a multi-modal large language model (MLLM). The method is suitable for efficient and accurate automatic classification of astronomical-to-star light variation curve images. The method comprises the following steps: firstly, extracting visual features of an optical variable curve image through a convolutional neural network, and embedding related physical parameters (such as period, amplitude and the like) into image data to form multi-modal input; then, image and text data are processed through a multi-modal large language model, and time domain features of a light variation curve and physical features of a variable star are learned, so that the accuracy of variable star classification is improved. The method is suitable for automatic classification of large-scale astronomical observation data, and especially has important application value in the fields of variable star classification, star feature recognition, intelligent image analysis and the like.

The utility model belongs to concrete production equipment technical field provides a kind of automatic concrete vibrating table, including vibrating table frame body, vibrating table structure unit, test mould fixed structure unit, platform cleaning scraping structure unit and waste classification processing unit. The automatic concrete vibrating table of the application, it has the function of vibrating, realizes the vibration and shakes solid to concrete forming test mould, cooperates test mould fixed structure unit and fixes concrete forming test mould, avoids the phenomenon that test mould slides and slurry splashes in the process of vibrating, simultaneously still has the function of cleaning and scraping, and can collect the concrete waste generated by vibration to waste classification processing unit, can automatically wash effectively maintain the clean of platform, and can realize the effective classification of waste.

This invention discloses a spring pressure testing device, comprising a base plate, a pressure testing mechanism, and a loading / unloading mechanism. Both the pressure testing mechanism and the loading / unloading mechanism are disposed on the upper surface of the base plate. The pressure testing mechanism includes a bottom support assembly and an upper pressure application assembly. The bottom support assembly is fixed to the upper surface of the base plate. The upper pressure application assembly includes a pressure testing section horizontally movable on the upper surface of the base plate and a horizontal drive section for driving the pressure testing section. The loading / unloading mechanism includes a spring moving section vertically lifted on the upper surface of the base plate and a vertical drive section for driving the spring moving section. The spring moving section is located directly above the bottom support assembly. The spring moving section includes a moving bracket and a horizontal conveyor belt. The horizontal conveyor belt is disposed on the moving bracket and is used for horizontally conveying the springs. The conveyor belt body of the horizontal conveyor belt is provided with a plurality of evenly distributed spring clamping members. This invention eliminates the need for manual intervention during loading and unloading, effectively improving testing efficiency.

This invention provides an object classification method based on lightweight networks, comprising the following steps: constructing a lightweight network, acquiring and processing sample images, training the lightweight network using the processed sample images, obtaining a trained lightweight network, and using the trained lightweight network to extrapolate the image of the object to be identified: extracting multiple different image features by dilated convolution of the image of the object to be identified according to multiple preset dilation rates, then assigning weights to the image features using adaptive weighting and performing feature fusion, finally outputting the classification result. This invention provides an object classification method based on lightweight networks, solving the problem of low classification accuracy achieved by existing object classification techniques in low-resource environments.

The invention relates to the technical field of agricultural informatization and fine management, in particular to a morchella esculenta growth environment dynamic optimization method based on Internet of Things and data fusion. The invention discloses a morchella esculenta growth environment dynamic optimization method based on Internet of Things and data fusion, and the method comprises the following steps: S1, collecting control parameters and demand parameters of a morchella esculenta growth environment, and determining a moderate interval and a stress interval according to the control parameters and the demand parameters; s2, obtaining a moderate interval association pair and a stress interval similar pair based on the moderate interval and the stress interval, and obtaining a moderate interval line segment set and a stress interval line segment set according to the moderate interval association pair and the stress interval similar pair; according to the method, the morchella esculenta growth rhythm is excavated, a critical point and a mutation point are accurately recognized, and a dynamic regulation and control plan is formulated accordingly; and through multi-stage data fusion, biological signals and interference are effectively distinguished, so that the environment regulation and control precision and the self-adaptive capability of the system are improved, and the defects of traditional static control are overcome.

The invention discloses a method for searching and identifying road diseases by using persistent coherence, and relates to the technical field of road defect detection. According to a topological data analysis method based on persistent coherence, path finding is carried out by using persistent coherence. Comprising the following steps: step 1, zooming an image again; 2, finding out a topological structure by using persistent coherence; step 3, bonding the area near the position; and 4, analyzing the shapes and sizes of different segmented regions to detect and identify the road disease. According to the topological fingerprint generation method based on continuous coherence, a novel and powerful technical path is provided for solving the essential feature extraction problem of a two-dimensional object, and path diseases can be found and recognized more accurately.

The utility model provides a cutting mechanism capable of quickly cutting polishing abrasive paper, which relates to the technical field of cutting and comprises a group of fixing blocks, a first motor is arranged on one side of the group of first fixing blocks, and first rolling shafts are movably inserted into the inner surface walls of the group of first fixing blocks. Second fixing blocks are installed on the outer surface walls of the first fixing blocks, a second motor is installed on one side of each second fixing block, second rolling shafts are movably connected to the inner surface walls of the second fixing blocks in an inserted mode, and a storage plate is installed above the first rolling shafts and the second rolling shafts; the polishing abrasive paper is conveyed through the two sets of rolling shafts, and the problems that in the production process of many abrasive paper, a semi-automatic cutting machine is relied on, a large amount of manual operation is needed, the manual operation efficiency is low, the production cost is high, and the like are solved. And the labor intensity and the operation hidden danger are increased.

This utility model discloses an aquatic seedling screening device, belonging to the technical field of screening devices. The aquatic seedling screening device includes a support frame and multiple sets of discharge hoppers disposed in the middle of the support frame. A surrounding plate is fixedly connected to the top of each discharge hopper. The device also includes: a screen disposed at the top of the support frame, with a vibration assembly fixedly connected to the bottom of the screen. The vibration assembly includes a spring telescopic rod, a U-shaped block, a connecting rod, and a rotating shaft, with the other end of the connecting rod rotatably connected to the connecting shaft; two sets of horizontal plates disposed at the bottom of the support frame, with a drive assembly fixedly connected to one side of each horizontal plate. The drive assembly includes a slide rail, a sliding plate, and a connecting frame, with two sets of rollers rotatably connected to both sides of the connecting frame; and mounting plates disposed on both sides of the support frame, with a slag removal assembly rotatably connected to the surface of each mounting plate. This utility model, through the arrangement of the vibration assembly and drive assembly, effectively avoids screen clogging, reduces maintenance costs, improves screening efficiency, and enhances the sustainability of the screening device.

This application proposes a model training method, a printing defect detection method, and related apparatus, relating to the field of model training and inference technology. The model training method includes: acquiring a defect image of a printed product and a first defect mask corresponding to the defect image; performing target processing on the defect image to obtain a defect sample image; wherein the target processing is used to mimic imaging interference factors in the defect detection image of the printed product; and training a target model to be trained based on the defect sample image, the first defect mask, and a standard template image of the printed product, wherein the target model is used to detect printing defects in the printed product, and the standard template image is used as a comparison reference image. The technical solution provided by this application can solve the problem of high false detection rates in traditional defect detection methods in the prior art.

The utility model relates to a kind of waste alloy material classification device based on LIBS technology, belong to waste alloy classification technical field, including conveying system, the end of conveying system includes multiple parallelly arranged transmission channel corresponding to different alloy categories;Conveying system is sequentially provided with cleaning system, LIBS detection system and classification system along transmission path, and the classification system is electrically connected with LIBS detection system.The classification system is set downstream in the end of the conveying system, and the classified material after identification is classified and transported to the corresponding channel, specifically including mechanical arm and the classification execution mechanism connected with the mechanical arm.The utility model is washed to the surface of waste alloy material before classification, improve the accuracy of detection;By setting LIBS detection system, reduce the detection time, ensure the accuracy and operating efficiency of classification system.

The invention discloses a meta-learning-based knowledge graph small sample node classification method, which is used for realizing efficient and stable small sample node classification with generalization ability on graph structure data. Firstly, a task difficulty-driven meta-training scheduling mechanism is provided, and a model is guided to gradually learn from easy to difficult by evaluating task complexity and dynamically sequencing and scheduling training tasks, so that training stability and generalization performance are improved. Secondly, a Poisson learning pseudo-label generation method based on a graph structure is designed, high-confidence pseudo-labels are generated by using unlabeled nodes and adjacency relations thereof, and KL divergence constraint optimization inter-class prototype distribution is introduced, so that the adaptability of a model to unseen classes is enhanced. And finally, a discriminative negative sample enhancement strategy is constructed, and the recognition capability of the model at the category boundary is enhanced by generating a discriminative negative sample pair in the feature space, so that the prototype confusion problem is effectively relieved.

The application discloses a virus-host RNA sequence classification method and device based on AI, and relates to the field of biological detection.The method comprises the following steps: mapping preprocessed short read RNA sequences to a host genome twice, assembling short read RNA sequences which are not mapped to the host genome into continuous RNA sequences, and screening RNA sequences with a length greater than 1000bp from the continuous RNA sequences; and performing AI classification on the RNA sequences with a length greater than 1000bp to obtain virus RNA sequences.The application combines host filtering, rapid assembly and AI classification, can significantly reduce the calculation amount and hardware pressure, realizes efficient and accurate virus sequence classification, and can quickly distinguish unknown viruses.

This invention discloses a method for constructing a safety training courseware system for manufacturing enterprises, a computer device, and a computer-readable storage medium. The method systematically builds a safety training courseware framework, reserving corresponding training courseware slots for each business activity involved in the manufacturing enterprise, thus improving the relevance and effectiveness of safety training. Simultaneously, the system can autonomously collect missing training courseware based on existing resources, and employs a series of incentive mechanisms to encourage upload nodes to specifically supplement the system with missing training courseware, ensuring the integrity of the training courseware system. Each training courseware uploaded by an upload node can be automatically stored in the optimal location, improving the standardization of training courseware storage.