18results about How to "Guaranteed retention" patented technology

The invention provides a three-dimensional reconstruction method and system in a nuclear irradiation environment, and relates to the technical field of computer vision. According to the method, a Swin Transformer network is adopted to perform feature extraction processing on an original image containing radiation noise, a constructed three-dimensional reconstruction network adopts a U-Net architecture, reservation of detail features is ensured through an encoding-decoding process, after radiation noise processing is completed, an FFB6D network is introduced to directly process a nuclear radiation environment reconstruction image, and the method has the advantages that the method is simple and convenient to operate and high in practicability. The accuracy and the stability of pose estimation under severe visual conditions are improved, and a geometric basis is provided for three-dimensional reconstruction. Model selection and optimization are carried out on all the modules adopted by the system according to the characteristics of the nuclear irradiation environment, technical resultant force is formed, a high-performance and full-automatic three-dimensional reconstruction system specially designed for the extremely severe visual environment such as nuclear irradiation is finally constructed, and the precision, robustness and practicability of three-dimensional perception in the field are remarkably improved.

The application discloses a multifunctional injectable hydrogel and a preparation method and application thereof, and relates to the technical field of multifunctional hydrogel preparation and application. In the preparation, CMC-AGE and OD-SH are prepared respectively, and DSPE-se-se-PEG@AC NPs nanoparticles are prepared, then the multifunctional injectable hydrogel is prepared by forming a hydrogel through CMC-AGE and OD-SH and adding DSPE-se-se-PEG@AC NPs nanoparticles and / or beta nerve growth factor. The application also includes the multifunctional injectable hydrogel prepared by the above method and the application thereof. The application constructs an in-situ hydrogel delivery platform with good biocompatibility, is used for co-delivery of DSPE-se-se-PEG@AC NPs and NGF, is used in ischemic stroke treatment, and reshapes the pathological microenvironment of a cerebral infarction area, so that cerebral ischemic injury is reduced.

The present application relates to the biological field, and particularly to a weak acid buffer and its use in an immunopeptidome separation reagent, the weak acid buffer comprising amino acids and / or polypeptides, the mass-to-charge ratio of the amino acids and / or polypeptides being theoretically predictable and experimentally reproducible, or the mass-to-charge ratio being beyond the preset acquisition range of a mass spectrometer; the weak acid buffer further comprising a volatile acid, the volume concentration of the volatile acid being 0.01% to 10%; and the pH of the weak acid buffer being 3.0 to 3.5. The weak acid buffer of the present application has high compatibility, can maximally avoid pollution and signal interference to an analysis system, has a simple experimental procedure, and the overall operation can be completed within tens of minutes, thus having low technical threshold and high acceptability; can improve the number of immunopeptide groups identified, thus improving the efficiency of the experiment; and the reagent is cheap and easy to obtain, thus the method has low cost, is suitable for large-scale or even industrial use, has high universality, and the like.

The invention discloses a fault record storage device, an upper computer and a fault record query system, and the device comprises a plurality of first communication interfaces which have different communication protocols and are used for accessing target monitoring equipment so as to obtain a plurality of fault records of the target monitoring equipment based on the corresponding communication protocols; the data processing unit is connected with the plurality of first communication interfaces and is used for storing a plurality of fault records and circularly covering the plurality of fault records based on a storage sequence of the plurality of fault records under the condition that the residual storage space of the data processing unit is smaller than a preset threshold value; the second communication interface is connected with the data processing unit and used for accessing an upper computer so as to send the data messages corresponding to the multiple fault records to the upper computer for a user to inquire, effective storage of the fault records can be achieved only through a single device, then the cost of the fault records can be reduced, and the user experience is improved. And the applicability of the fault record storage device is improved.

This invention relates to a water purification agent with post-adjusted basicity, its preparation method, and its application, belonging to the field of water purification agent technology. The invention provides a method for preparing a water purification agent with post-adjusted basicity, the specific steps of which are: adding a regulator to a water purification agent to obtain a water purification agent with post-adjusted basicity; wherein the water purification agent is an iron salt solution, specifically FeCl3 solution, and the regulator is aluminum citrate solution; or the water purification agent is an aluminum salt solution, and the regulator is at least one of an organic acid and dilute hydrochloric acid. This invention, through the method of post-adjusting basicity, changes the basicity of the water purification agent system, thereby improving the utilization rate of the water purification agent. The prepared water purification agent with post-adjusted basicity ensures water quality treatment effect and stability, and reduces the dosage, thereby reducing the cost of the reagent. In high-alkalinity water bodies, the hydrolysis rate decreases by 50%, and the reagent utilization rate increases to 88%.

The utility model discloses a dental spherical attachment negative part replacing device which comprises a connecting rod, a threaded operating handle is arranged at one end of the connecting rod, a ball head is arranged at the top end of the threaded operating handle, the diameter of the ball head is 0.1-0.2 mm larger than the inner diameter of a spherical cap of a dental spherical attachment negative part, and the connecting rod is connected with the connecting rod. The ball head and the threaded operating handle are in smooth transition through an arc chamfer, the ball head and the threaded operating handle are made of stainless steel materials, the surface roughness of the ball head is Ra0.4-Ra0.6 micrometers, and the length of the threaded operating handle is 3-4 cm. The spherical cap is reasonable in structural design, due to the design of the ball head structure, possible damage to the internal structure of the spherical cap when the spherical cap is replaced is avoided, the spherical cap can be rapidly and accurately placed in the clamping groove, needing to be replaced, of the spherical attachment female part, the retention force of the spherical cap is better guaranteed, the replacement time is greatly shortened, and time and labor are saved.

The embodiment of the present disclosure provides an image restoration method based on a text-guided and detail-preserving diffusion model, which belongs to the technical field of image processing, and specifically comprises the following steps: obtaining a preliminary degraded image according to a clear image, encoding the clear image into a latent space through an encoder and adding noise to obtain a noise latent variable; inputting the preliminary degraded image and the noise latent variable into a conditional diffusion model as conditions to predict noise added in the t-th step, calculating a latent variable that preserves details and clarity, and decoding a new degraded image through a decoder according to the latent variable; inputting the new degraded image and the noise latent variable into the conditional diffusion model as conditions again to obtain a final predicted noise, and calculating a loss according to the final predicted noise and a preliminary predicted noise to train the conditional diffusion model; saving the model weight of the trained conditional diffusion model, and completing the image restoration process of a target image by using the inference process of DDIM. Through the scheme of the present disclosure, the restoration efficiency and adaptability are improved.

This utility model discloses a dispensing mechanism for a cold drink machine, including a dispensing tray, a housing, and a handle. The dispensing tray is installed on the housing, and the handle is installed on the dispensing tray. The dispensing tray has a second dispensing port, and the bottom of the dispensing tray has a dispensing cover with a first dispensing port connected to the second dispensing port. The bottom of the handle has a sealing element and a transmission assembly. The transmission assembly connects the handle and the sealing element, and is used to control the movement of the sealing element. The sealing element is used to control the opening and closing of the second dispensing port. In this utility model, ice cream or smoothies enter the dispensing cover from the second dispensing port and then exit from the first dispensing port at the bottom of the dispensing cover, effectively avoiding the problem of ice cream or smoothies splashing everywhere when pushed out of the freezing cylinder in traditional cold drink machines.

The application discloses a principal component analysis multi-bias interaction-based longitudinal federal learning optimization method, an electronic device and a storage medium, and belongs to the technical field of privacy calculation. In order to improve the efficiency of a longitudinal federal neural network model in the case that the capacity of a data set is small, the application trains participating parties including a training initiator and a training assistant, trains data of the training initiator and data of the training assistant by using a forward propagation method, obtains a training result of the forward propagation method, trains the training result of the forward propagation method by using a backward propagation method, and the training assistant and the training initiator respectively update model parameters, so that one round of training is completed. The training is repeated until the training result reaches a precision requirement or a stop condition, and neural network training in a longitudinal federal scene based on multi-bias interaction is completed. The principal component analysis data dimension reduction method is used, so that the filtering function of features is guided by multi-party data information, and the result is more persuasive.

The invention relates to the technical field of machine vision, in particular to a ton bag cargo robust visual detection system based on multi-cluster feature fusion and attitude correction driving. The system comprises a non-local mean value enhancement module, a multi-feature clustering fusion module, a geometric feature extraction module, a dominant attitude clustering module, an attitude correction transformation module and a cargo segmentation judgment module, and non-local mean value filtering is carried out on an original image of the ton bag cargo to obtain an enhanced image; according to the method, the non-local mean value enhancement module and the multi-feature clustering fusion module are carried, efficient enhancement and multi-dimensional feature fusion of the ton bag cargo image can be realized, the fusion segmentation image is generated through spatial alignment weighted stacking of the color clustering cluster and the texture clustering cluster, feature representation of the homogeneous region of the ton bag cargo is more comprehensive and accurate, and the method is more suitable for mass production. The integrity and effectiveness of geometric feature extraction are effectively improved, and the feature analysis capability of the system on the ton bag cargo image is remarkably improved.

This invention discloses a method, apparatus, and storage medium for managing electronic project documentation in investment banking. First, it manages the project documentation directory template based on configurable and version management; then, it binds each stage of the project documentation directory; next, it maintains or rebuilds the project documentation directory; it stores project documentation files; it locks the project documentation directory; and finally, it submits the project documentation for verification. This invention not only enables standardized and customized flexible changes to the project documentation directory but also effectively assists in controlling the correctness and integrity of the project execution process. The coordinated operation of directory locking, documentation storage, and documentation submission effectively controls the execution process, ensuring its correctness and integrity. This invention guarantees the flexibility of investment banking project documentation to adapt to internal and external changes and the standardization of documentation file maintenance, thereby improving the efficiency of investment banking operations and the ability to control risks.

The application discloses a multi-modal medical recommendation method based on a double-layer gating mechanism and related devices, and relates to the technical field of artificial intelligence medical treatment. The method obtains text data, numerical data and category data of a patient, and pre-processes the data to obtain a data set; the data set is input into a multi-task learning model for training until a training round is reached or a preset condition is met, a trained multi-task learning model is obtained, the multi-task learning model extracts features from the text data, the numerical data and the category data through a text feature branch, a numerical feature branch and a category feature branch respectively to obtain text features, numerical features and category features; the text features, the numerical features and the category features are simultaneously input into a double-layer gating network for mode-level gating and feature-level gating processing to obtain final feature representations; and the final feature representations are input into a multi-task head network to obtain classification prediction results of a plurality of tasks. The application can provide accurate personalized medical recommendations for patients.

The invention relates to the technical field of hydropower station site selection, and discloses a hydropower station site selection method and system based on GIS spatial analysis, and the method employs a natural node splitting strategy to split a river reach, calculates the runoff volume, guarantees the reservation of a river reach structure, and optimizes the water resource configuration. The method improves the reasonability and resource utilization efficiency of hydropower station site selection, deduces the layout scheme of cascade hydropower stations through automatic calculation and multi-round optimization in combination with multi-source data such as terrain, hydrology, geology, environmental constraints and the like, ensures efficient utilization of water resources and coordinated operation of a hydropower system, and generates river planning scheme longitudinal profile drawings of different river reaches. According to the method, the problems of errors, scattered point selection results and low visualization degree can be reduced, the site selection and planning efficiency can be greatly improved, a scientific and accurate solution is provided for hydropower development, and sustainable energy utilization is promoted.

The invention relates to the technical field of health food, and discloses a polygonatum sibiricum, spina date seed and lily composition for improving sleep quality and application thereof.The composition is prepared from, by weight, 30-50 parts of processed polygonatum sibiricum extract, 20-40 parts of fried spina date seed extract, 15-30 parts of lily extract, 5-15 parts of gamma-aminobutyric acid and 1-5 parts of melatonin. And auxiliary materials acceptable in pharmacy or food science. The processed rhizoma polygonati extract plays a role in consolidating basis, the fried semen ziziphi spinosae extract is longer than nourishing heart and tonifying liver, and the lily extract plays a role in clearing away the heart fire and soothing the nerves for sleep disorder caused by restlessness; meanwhile, gamma-aminobutyric acid and melatonin are scientifically introduced into the formula, the gamma-aminobutyric acid relieves emotion by adjusting a central nervous inhibition system, and the melatonin supplements endogenous melatonin to adjust a sleep-awakening period and forms a multi-target and multi-layer synergistic effect with traditional extract components, so that the sleep disorders of different reasons are comprehensively improved.

The application relates to a planetary gearbox fault diagnosis algorithm automatic generation system and method based on a large language model, which comprises the following modules: a large language model generation module, which is used for receiving prompt words, selecting parent planetary gearbox fault diagnosis algorithm and architecture component knowledge, and combining with the characteristics of the planetary gearbox fault diagnosis task to call the large language model to generate new planetary gearbox fault diagnosis algorithm code; a dynamic code execution and verification module, which is used for code execution and verification; an architecture evaluator module, which is used for training and performance evaluation; an intelligent evolution strategy module, which is used for analyzing the training history and obtaining an analysis result; a program database module, which is used for storing and managing the population of the planetary gearbox fault diagnosis algorithm and constructing prompt words; and an evolution control engine, which is used for coordinating and controlling the work processes of the modules and controlling the number of iterative evolutions to realize evolution circulation. Compared with the prior art, the application has the advantages of automatic generation, iterative evolution and the like.

The present application relates to the technical field of artificial intelligence model optimization, in particular to a lightweight construction method of a multi-task AI prediction model in the field of low-altitude economic aviation, comprising: obtaining the historical execution frequency proportion of a plurality of sub-tasks in the model, and calculating the complexity entropy value of each sub-task based on the frequency proportion; for the to-be-compressed parameters of the model, combining the static weight amplitude and the dynamic gradient sensitivity, calculating the importance score of each to-be-compressed parameter; combining the entropy value and the importance score, generating a dynamic compression threshold; and structurally pruning the to-be-compressed parameters with an importance score lower than the dynamic compression threshold from the model to obtain a lightweight model. The present application solves the technical problem that the real-time change of task complexity cannot be adapted due to the use of a fixed compression threshold, resulting in resource waste in simple task scenarios and precision damage in complex task scenarios.

Embodiments of the present specification provide a video processing model training method and a video question answering method, wherein the video processing model training method comprises: obtaining a sample video, and sample target object question text and sample target object description text corresponding to the sample video; generating video feature information and target object question text feature information based on the sample video and the sample target object question text, and inputting the video feature information and the target object question text feature information into a feature compression module in a video processing model to generate video compression feature information; generating predicted target object description text based on the video compression feature information and the text feature information; and training the video processing model based on the sample target object description text and the predicted target object description text. By obtaining sample videos and text information related to a target object in the sample videos, the model including the feature compression module is trained to improve the understanding ability and processing efficiency of the video processing model for the target object.

This invention belongs to the field of towel manufacturing technology, specifically a finishing process for long-staple cotton towels, including yarn treatment, fabric weaving, and dyeing. In this invention, after the long-staple cotton is produced into yarn, the yarn undergoes a preliminary dyeing process. The purpose of this is to create a clear color difference between the knotted areas and the normal areas. During subsequent weaving, when visually inspecting the fabric, the significant color difference allows for more accurate identification of the knotted areas. This enables precise napping treatment of the knotted areas in subsequent processing, improving the quality of the napping process. This reduces the impact of the knotted area on the quality of the finished towel, and can even prevent the appearance of white spots, ensuring the quality and comfort of the towel.