99results about How to "Easy to classify" patented technology

The invention discloses an intelligent suspension conveying device for building material production, which comprises a cable rod, a traversing shaft is arranged in the cable rod, a lifting cable rack is arranged at the bottom of the traversing shaft, a maintaining mechanism is arranged in the lifting cable rack, and an anti-collision mechanism is arranged at the bottom of the maintaining mechanism. The auxiliary rotating shaft is rotationally connected to the inner wall of the lifting cable frame, the maintaining cable is wound around the outer circumferential face of the auxiliary rotating shaft, the main holding cable is wound around the outer circumferential face of the main rotating shaft, and the lifting cable frame is slidably connected with the inner wall of the transverse moving shaft. Therefore, the stroke weight of the host cable is dispersed, shaking of the pipeline in the transportation process is limited, and the stability and safety in the transportation process are improved.

The invention provides a split-phase cooperative asphalt mixture digital image multi-component segmentation method, and relates to the field of asphalt mixture image segmentation, and the method comprises the steps: obtaining a digital image of an asphalt mixture section, converting the digital image into a gray image, and carrying out the preprocessing of the gray image, and obtaining a to-be-segmented image; performing gap segmentation on the to-be-segmented image to obtain a gap phase image mask; segmenting the to-be-segmented image by using the RAN-UNet deep learning image segmentation model to obtain an aggregate image mask; performing three-dimensional voxel reconstruction to obtain a three-dimensional digital model, and performing connectivity analysis and edge detection to obtain a connected gap mask, an open gap mask and a closed gap mask; performing critical particle size screening and structural gap identification to obtain a structural gap mask and a mortar internal gap; and carrying out particle size screening on the aggregate image mask, extracting an asphalt mortar region, and slicing and converting the asphalt mortar region again to obtain a mortar phase image. According to the invention, the segmentation precision of the asphalt mixture digital image is improved.

The present application relates to the field of chip package testing, in particular to a chip package testing system and a testing method thereof. The system comprises a testing box, the inside of which is divided into a testing room and a chip in-out room. The testing room is divided into a feeding room with adsorption equipment and detection equipment, a discharging room, and two groups of detection rooms. The rotating plate is divided into four groups of rotatable detection stations, each of which is provided with a positioning assembly. The testing box is also provided with a feeding box, a qualified product discharging box, an unqualified product discharging box, and a controller. Through the rotation of the detection stations, the feeding room, the two groups of detection rooms, and the discharging room are sequentially opposite, and the feeding, detection, and classified discharging are automatically performed. The structure is compact, and the items for simultaneous detection can be installed and replaced according to needs. The detection process is cyclic, continuous, efficient, and intelligent. After the detection is completed, the qualified products and unqualified products are classified and recycled, the detection results are intuitive, the classification and statistics are convenient, and the detection efficiency is improved.

The application relates to the field of optical fiber sensing, and more particularly to an event detection method, system and device for optical cable detection and a storage medium. The method comprises the following steps: inputting optical fiber data of an optical cable into a first processing unit for processing to obtain space-time features; inputting the optical fiber data into a second processing unit for processing to obtain frequency domain features; inputting the space-time features and the frequency domain features into an attention processing unit after splicing for feature enhancement to obtain enhanced features; inputting the enhanced features into a bottleneck layer processing unit for processing to obtain bottleneck features; wherein the bottleneck layer processing unit comprises at least two bottleneck layer processing subunits, and the bottleneck layer processing subunits can perform feature transformation processing on the enhanced features; and inputting the bottleneck features into a backbone network for optical fiber event detection to obtain an optical fiber event detection result corresponding to the optical fiber data. The application can improve the accuracy of optical fiber event detection in an optical cable.

The application relates to a wireless communication same-frequency interference detection method based on a CNN-SVM and a particle swarm optimization, which comprises the following steps: performing short-time Fourier transform on a received baseband signal to generate a log power spectrum time-frequency graph; automatically extracting a deep space feature vector through a convolutional neural network; performing global self-adaptive optimization on a penalty parameter and a kernel parameter of a support vector machine through a particle swarm optimization algorithm; and training the support vector machine with optimal parameters to perform binary classification judgment on a to-be-detected feature. Through a triple coordination mechanism, the application overcomes the problems of fixed parameters, poor robustness and the need for a large amount of labeled data in traditional methods, and realizes high detection probability and low false alarm rate under different signal-to-interference-and-noise ratios and bandwidth conditions.

The invention discloses an unmanned aerial vehicle remote sensing tree species classification method and system fusing a cascade canopy attention mechanism, and the method comprises the steps: obtaining unmanned aerial vehicle RGB image data of a campus scene, carrying out the preprocessing of the data, and obtaining a standardized input image; constructing a classification model fusing the cascaded canopy attention mechanism, wherein the classification model comprises a background suppression module, a fine-grained feature guide module, a channel alignment neck module and a global-local fusion classification head; inputting the standardized input image into a classification model, and obtaining an enhanced canopy image through a background suppression module; inputting the enhanced canopy image into a module carrying fine-grained feature guidance, and extracting fine-grained canopy features; performing channel calibration and feature smoothing on the fine-grained canopy features through a channel alignment neck module to obtain feature mapping of a unified dimension; and fusing the global canopy semantics and the local texture structure in the feature mapping by using a global-local fusion classification head, and outputting a tree species classification result.

The utility model relates to the technical field of storage bag, disclose a kind of portable storage bag, including storage bag body, the inner side edge position of storage bag body is provided with first zipper, the inner side wall of storage bag body is provided with external closure layer, the inner side wall of storage bag body is fixedly connected with connecting layer, the inner side wall of storage bag body is fixedly connected with split layer, the outer surface of storage bag body is fixedly connected with support air bag column, the upper surface of storage bag body is fixedly connected with elastic air bag, the side of elastic air bag is provided with air inlet, the other side of elastic air bag is provided with check valve, the one end of check valve is fixedly connected with exhaust valve, the utility model is set to external closure layer, second zipper, connecting layer, third zipper and split layer, and by the setting of support air bag column, elastic air bag, air inlet, check valve and exhaust valve, effectively solve the problem that the internal article of existing storage bag is easy to be chaotic, and the structure is insufficiently supported.

The application provides a kind of foodborne microorganism identification method based on metal omics coupling XGBoost algorithm, belongs to microorganism identification technical field, the foodborne microorganism identification method includes the following steps: 1) with foodborne microorganism strain as research object, ICP-MS is used to analyze each standard strain metal group characteristics, combined with XGBoost algorithm to construct foodborne microorganism classification model;2) the ICP-MS full quantitative analysis is used to detect the bacterial species, and the detection data of the metal elements of the bacterial species to be detected are obtained;3) the obtained detection data is substituted into the obtained foodborne microorganism classification model for analysis, and the bacterial species to be detected is identified.The foodborne microorganism classification model constructed by the application can quickly and accurately identify 18 kinds of foodborne microorganisms, and lays a good foundation for the broad-spectrum and efficient classification and identification of more kinds of foodborne microorganisms in the future.

This invention relates to the field of chip inspection technology, and more particularly to a chip non-destructive testing system based on embedded current scanning. The system includes a current scanning module, an information acquisition module, a data analysis module, an abnormal scanning region generation module, and a scanning path generation module. Instead of simply performing a higher-density secondary scan of the entire chip, this invention intelligently filters out abnormal scanning regions requiring special attention based on the initial scan results and calculates their scanning priority. This allows valuable inspection time and computing resources to be concentrated on the most suspicious and severe areas, avoiding unnecessary repeated scanning in normal areas. The system dynamically selects the optimal scanning path based on the regularity of the region shape; for example, parallel lines are used for regular regions, and radial lines are used for irregular regions. This makes the scanning path more closely match the geometric features of the defect, obtaining a more complete region characterization with fewer scanning points, resulting in higher path planning efficiency.

The invention discloses background layer updating during coding. A method performed by an image processing apparatus for updating a background layer during encoding of a scene is provided. A scene is encoded based on classifying objects depicted in the scene as foreground or background. The background is divided into ranked background layers, wherein each background layer is associated with a respective depth model. The method comprises detecting (S102) a change in an image portion of a background layer. The method comprises calculating (S104) a difference between the image portion and a corresponding image portion of a background layer ranked behind the one background layer. The method comprises selecting (S106a) the background layer ranked behind the one background layer to represent the image portion when the difference is less than a threshold.

The invention belongs to the technical field of wind power generation, and particularly relates to a wind turbine generator state evaluation method and a related device, and the method comprises the steps: obtaining an electromagnetic transient response signal of a to-be-evaluated wind turbine generator; performing empirical mode decomposition on the electromagnetic transient response signal of the wind turbine generator to be evaluated to obtain a plurality of image forming model components; performing feature extraction on each image forming model component, and performing normalization processing to obtain a feature vector; classifying the feature vectors by using a pre-constructed state classification model to obtain a state evaluation result of the to-be-evaluated wind turbine generator; wherein the pre-constructed state classification model is a pre-trained support vector machine; according to the method, the fault features in the electromagnetic transient response signals of the wind turbine generator can be effectively extracted, and the accuracy and robustness of state evaluation are improved; meanwhile, non-linear and non-stationary signals can be decomposed in a self-adaptive mode, the time domain features of the signals are reserved, and the information amount and accuracy of state evaluation are improved.

The utility model belongs to the technical field of table tennis ball quality inspection, and provides a discharging mechanism of table tennis ball quality inspection equipment, which comprises a discharging driving device, a discharging sucker mechanism and a discharging collecting mechanism, and the discharging driving device and the discharging collecting mechanism are arranged on an equipment frame body of the table tennis ball quality inspection equipment. The driving end of the discharging driving device is connected with the discharging suction cup mechanism, the discharging driving device drives the discharging suction cup mechanism to move relative to the equipment frame body, and the discharging collecting mechanism is oppositely arranged below the discharging suction cup mechanism. The table tennis ball quality inspection device is reasonable in structural design, table tennis balls in a quality inspection placement area can be automatically transferred and collected, manual participation is reduced, and the automation degree is higher.

The application discloses a radioactive substance packaging and sorting equipment, which comprises a packaging device and a sorting device. The packaging device comprises an encapsulating part configured to accommodate and encapsulate radioactive substances to form sealed packages. The sorting device comprises a sorting part configured to receive the sealed packages from the packaging device and analyze the content of the radioactive substances in the sealed packages, so as to place the sealed packages in different storage boxes according to the content of the radioactive substances in the sealed packages. The radioactive waste can be classified and treated according to the content of the radioactive substances in the sealed packages, the disposal amount of the radioactive waste is effectively reduced, and the disposal cost of the radioactive waste is saved.

The embodiment of the application relates to an algorithm and a device of a Bayesian classifier of a non-uniform background, which are applied to an underwater active sonar system, the algorithm comprising: obtaining underwater data to be measured and auxiliary data through the active sonar system; the number K of the auxiliary data is greater than 0; modeling classification of the unknown covariance matrix structure into a binary hypothesis testing problem; hypotheses of the binary hypothesis testing problem comprise H i Wherein i=0, 1, H0 is a case that the unknown covariance matrix is a complex conjugate symmetric matrix; H1 is a case that the unknown covariance matrix is a real symmetric matrix; a Bayesian model is set, the Bayesian model comprising a complex inverse Wishart random matrix and a real inverse Wishart random matrix; a classifier for distinguishing the two hypotheses is obtained by using a minimum error probability criterion.

The application provides a cotton dyeing effluent water quality prediction method and device and application, belongs to the printing and dyeing technical field, can predict the cotton dyeing effluent water quality, obtains effluent water quality parameters of multiple sections of the cotton dyeing process to be operated in a future period of time, so that the cotton dyeing effluent can be classified and treated, the requirements of each process link on the water quality of recycled water are met, the treatment efficiency of the cotton dyeing process effluent is improved, and the recycling cost is reduced. The method inputs process parameters and dye parameters of the cotton dyeing process to be operated into an effluent water quality prediction model, and the effluent water quality prediction model outputs effluent water quality parameters of the cotton dyeing process to be operated.

The utility model relates to food storage fresh -keeping device technical field, concretely is a kind of food processing storage fresh -keeping device, including fresh -keeping box, the upper portion of fresh-keeping box is provided with upper storage box, the lower portion of fresh-keeping box is provided with lower storage box, the inside sliding connection of upper storage box has N type frame, the inside of N type frame is provided with storage box, the inner chamber top of lower storage box is provided with partition, the bottom of partition is provided with corrugated partition pipe, this food processing storage fresh -keeping device, by the inside sliding connection of upper storage box N type frame, N type frame is pulled out the inside of upper storage box, the storage box embedded in the inside of N type frame can be expanded to both ends under the fixed effect of sliding block, it is convenient to place and take food, the partition set in the inside of storage box can reduce the damage due to accumulation in the storage process of food, improve the convenience and fresh -keeping effect of food storage.

This utility model discloses a waste collection device for flat round tube production, belonging to the technical field of waste collection equipment. It includes a tank, and a separation mechanism and a compression mechanism installed on the tank. The separation mechanism uses a handheld vacuum cleaner to suck up the waste, separating dust and debris through an inclined filter screen inside a waste box. The dust falls to the bottom of the box, and a striking rod strikes the filter screen, causing the debris to slide into the tank. The compression mechanism drives a pressing block via an electric push rod, compressing the debris and scraps inside the tank into blocks under the guidance of a guide rod. This utility model achieves automatic separation and compression of waste, solving the problems of mixed waste collection, scattered waste collection difficulties, and inconvenient transportation. It has the beneficial effects of clear classification and efficient collection.

The utility model discloses a vehicle -mounted refrigerator with bottle body support, include: refrigerator main part has the shell and the inner container and the upper cover, refrigeration component is used for refrigerating and heating for refrigerator, bottle body support subassembly, including support main part, support main part has the bearing plate and the positioning board, bearing plate with support main part is connected, the positioning board can overturn installation, has the positioning hole for limiting bottle body. The utility model discloses beneficial effect lies in, realizes vehicle -mounted refrigerator refrigeration and heating, and has bottle body support subassembly simultaneously, guarantees that bottled article will not be damaged in the vehicle driving process, satisfies a variety of use needs.

The utility model provides a kind of fusing body resistance full detection test equipment, it is related to fusing body resistance detection field, and it includes: base mechanism, the base mechanism includes support base, unqualified collection box and tester, the top of the base mechanism is provided with sorting mechanism, the sorting mechanism includes support frame, good product collection box, slide rail frame, finger cylinder and clamping claw, the top of the base mechanism is provided with detection mechanism, and the detection mechanism includes fixed frame, rotating motor, rotating disc, positioning fixture, product body to be measured, cylinder frame, drive cylinder, test probe and inductor.The utility model, the rotation of rotating disc makes that product body to be measured can rotate to the below of test probe, so that detection staff only needs to place product body to be measured to the top of positioning fixture, reduces the work pressure of staff, while it is convenient to classify product body to be measured, and it is convenient to sort after detection of product body to be measured.

The utility model discloses a kind of cleaning and wrong shelf detection equipment for library, it is related to library equipment maintenance technical field, comprising: equipment main body, mounting arm, dust suction member, dust suction fan, pipeline, detection device, mounting arm is located in equipment main body;Dust suction member is located in mounting arm and is communicated with dust suction fan by pipeline, dust suction fan is located in equipment main body;Detection device is equipped on equipment main body and / or mounting arm, detection device is configured as the label information of book can be identified and sends label information to host computer. Thus, by setting dust suction member and dust suction fan, dust on bookshelf and book surface can be cleaned, cleaning efficiency is high, cleaning effect is better, and, by setting detection device, books placed in wrong bookshelf can be detected while cleaning bookshelf and book surface dust, so that staff can classify books placed in wrong bookshelf, save manpower and time cost, improve the maintenance level of library.

The utility model relates to the technical field of water-quenched slag treatment, and discloses a screening device for ironmaking water-quenched slag treatment, which is characterized in that the upper surface of a box body is communicated with a feeding hopper, the lower surface of the box body is communicated with a blanking hopper, two through holes are formed in the box body, and the inner surfaces of the two through holes are movably connected with a screening tray; the side surface of the box body is fixedly connected with an external shell, the external shell is located in front of any through hole, the inner surface of the external shell is fixedly connected with an electric push rod, the other end of the electric push rod is fixedly connected with a connecting piece, the connecting piece is movably connected with the sieve tray, the side surface of the sieve tray is fixedly connected with an elastic buckle, and the elastic buckle is movably connected with the connecting piece; by means of the components, large materials in water-quenched slag can be crushed, better separation and classification are facilitated, the crushed materials can be evenly scattered on the sieve tray, it is ensured that the materials are evenly distributed in the screening process, and therefore the screening accuracy and efficiency are improved.

The present invention refers to a composite material comprising a fire-proof resin and gelcoat. The present invention also relates to the method for producing a composite material and to the use of the composite material, in transport means, preferably railway, and in construction. The present invention also refers to a railway transport component that comprises the composite material.

The utility model relates to e -commerce goods packing technical field, specifically disclose e -commerce goods packing and transporting box, including the paper box for parcel target object and set up on the closing cover of paper box, the paper box includes the box, the inner wall both sides of box are provided with the bending part, the bottom of box is provided with the bottom plate, the recess that is linear array setting is established at bottom plate, is used for reinforcing bottom plate, the both sides of bending part are provided with the reinforcing rib no.

The present application relates to a hyperspectral band selection method based on KL divergence weighted sparse self-representation, belonging to the technical field of hyperspectral image processing. The method obtains a hyperspectral image cube, converts it into a two-dimensional spectral matrix; uses PCA and superpixel segmentation algorithm to segment the image into multiple regions; constructs a neighborhood graph of the spectral matrix, calculates a similarity graph and performs a normalization operation; measures the information quantity difference between bands based on KL divergence, calculates the dissimilarity matrix of the spectral matrix; constructs a weighted sparse self-representation model that fuses spatial-spectral information, solves the coefficient matrix; arranges the L2 norm of the coefficient matrix row vector in descending order, selects the bands corresponding to the first M row numbers as the final band subset. The method fuses the spatial-spectral information of the hyperspectral image, reduces the similarity of the selected bands based on KL divergence, effectively improves the accuracy of band selection, and is beneficial to subsequent classification.

The invention discloses a radar HRRP deformation target recognition method based on feature correction, and the method comprises the steps: carrying out the preprocessing of the radar one-dimensional HRRP echo data of a to-be-recognized target, and obtaining the HRRP original target data; superposing block sparse signals with different sparseness and Gaussian white noise on the HRRP original target data to obtain HRRP deformation target data; constructing a lightweight feature extraction network, and performing pre-training by using HRRP original target data; a feature correction network comprising a multi-dimensional parallel feature correction module, a feature fusion module and a de-noising post-processing module is constructed, and training is carried out after the feature correction network is cascaded with the feature extraction network; and combining the feature extraction network with the feature correction network to obtain a deformation target identification model so as to realize identification and classification of a deformation target. According to the method, the intrinsic feature distribution of the deformed target can be effectively recovered, and the recognition accuracy and generalization ability of the non-cooperative target in a low signal-to-noise ratio environment are remarkably improved.

The invention relates to the technical field of image recognition, in particular to a few-sample image classification method for embedding feature reconstruction uncertainty, which comprises the following steps: acquiring a few-sample image data set; constructing a local feature extraction model fusing an attention mechanism and a convolutional neural network, and obtaining image local features and weights thereof based on the model; designing a feature reconstruction model based on a Gaussian process, reconstructing image features of the query set, and obtaining a corresponding reconstructed image and uncertainty thereof; modeling the similarity between the query samples and the support samples, and classifying query images; constructing a fusion loss function containing classification loss and feature reconstruction loss, and optimizing the fusion loss function to obtain a final image classification model; processing a new few-sample image classification task by using the model, and classifying the image; according to the invention, a few-sample image classification method with embedded feature reconstruction uncertainty is realized, and the accuracy and reliability of few-sample image classification are effectively improved.

This invention provides a multi-channel continuous fast packet reception, sorting, and lock-free asynchronous read / write method, relating to the fields of computer communication and storage technology. The method includes: employing a dynamic expansion mechanism to continuously receive and store multi-channel data packets; classifying and sorting the stored data packets; and asynchronously reading and writing the parsed data packets in a lock-free manner. This invention generates an independent message stack for each channel, enabling continuous reception and storage of messages from multiple channels without pre-allocating excessive memory space. The speed of receiving buffer expansion is dynamically adjusted based on data flow rate. Fast classification and sorting of data packets is achieved using a hash table or sorted binary tree. A circular stack is created to sequentially store the parsed data content of each packet, updating the stack top position after data content is stored. When reading data, the starting position for reading is found using the stack top position and the number of packets read, and then the data is copied, thereby achieving lock-free asynchronous parallel data read / write.

This invention belongs to the field of data processing and artificial intelligence technology, specifically a method and system for automatic classification of medical device patents based on deep learning. The method includes the following steps: acquiring textual data of medical device patents and performing preprocessing; introducing medical device classification standards as an external knowledge base, and calibrating the labels of the preprocessed text data to form a standard dataset; for categories with insufficient sample numbers in the standard dataset, generating synthetic data based on the patent text using a generative model; constructing a deep learning classification model that integrates global and local semantic features; training the deep learning classification model using the enhanced dataset, and inputting the medical device patent text to be classified into the trained model for classification prediction, outputting its corresponding medical device classification standard category. This invention solves the problems of mismatch between existing patent classification systems and industry regulatory standards, and low classification accuracy.

The application discloses a model training method, a category prediction method, a device, a computing device and a computer readable storage medium, to solve the problem of low category prediction accuracy of the model. The training method comprises: training a category prediction first model by using an original sample object set; determining the matching degree between the label of the sample object and the predicted category of the sample object by using the trained first model; selecting sample objects meeting a preset condition from the original sample object set; and training a category prediction second model by using the selected sample objects. Since the sample objects used for training the category prediction second model are sample objects with relatively high label annotation accuracy, compared with the category prediction model trained directly based on the dirty data set, the category prediction model trained by the method provided in the application embodiment has higher accuracy.