20results about How to "Reduce perception" patented technology

The invention provides an end-to-end automatic driving system and method. The system comprises an environment sensing module, a feature extraction module, a data fusion module, a decision generation model module and a decision analysis module. The environment sensing module comprises a plurality of types of sensors, and each sensor collects environment information data of different modes. The feature extraction module extracts features of each modal environment information data. And the data fusion module fuses the features of the modal environment information data to obtain fused features. The decision generation model module determines a driving decision by using a trained decision generation model based on the fusion features. And the decision analysis module converts the driving decision output by the decision generation model module into a specific vehicle control instruction. The method has higher adaptability to complex and dynamic driving scenes, reduces the cost of data annotation during training, and improves the generalization ability of the model in various driving scenes.

The invention provides a digital watermark embedding and detecting method, equipment and a medium, and aims to solve the problems of easy perception, opaqueness and low robustness of the existing text digital watermark technology. The method comprises the steps that an Nth target clause is predicted, and at least two (N + 1) th candidate clauses are generated; according to the Nth target clause and a preset first watermark feature coding sequence, a first watermark feature coding value of an (N + 1) th target clause is determined, and the (N + 1) th target clause is the next clause of the Nth target clause; determining the (N + 1) th target clause according to the prediction probability of the at least two (N + 1) th candidate clauses, the first watermark feature coding value and a preset readability threshold; under the condition that the N + 1 target clauses comprise text end characters, a full text embedded with watermarks is determined, the full text embedded with watermarks comprises the N + 1 target clauses, and N is an integer larger than 0. According to the invention, low perception, transparency and robustness of digital watermark embedding can be realized.

The invention discloses a method for pushing data by combining a microfluidic system with a large model. The method comprises the following steps: step 1, carrying out diversity evaluation in a rearrangement stage of intelligent recommendation; and step 2, based on rule constraint, a heuristic algorithm or a depth model, performing intelligent recommendation rearrangement according to a diversity evaluation result, and performing accurate control by adopting a micro-fluidic chip built-in rule diversity scattering algorithm. Diversity regulation and control are carried out in the final link rearrangement stage of intelligent recommendation, diversity definition is matched with diversity perception of a user, various rule diversity scattering algorithms are built in by adopting a micro-fluidic chip technology, accurate control is carried out through a chip, damage to an original sequence is reduced, correlation can be reserved to the maximum extent, and the accuracy of recommendation is improved. The phenomenon that contents with the same attribute are gathered is avoided, and the use efficiency is high.

The invention relates to the technical field of fish oil processing for pet food. The processing method comprises the following steps: adding evening primrose oil and borage seed oil into fish oil, carrying out instantaneous high-temperature heating, cooling, adding a rosemary extract, uniformly mixing, adding a filter aid, stirring, winterizing, and filtering to obtain the fish oil with low fishy odor. The fish oil prepared by the method disclosed by the invention can effectively reduce the perception of people on fishy smell and peculiar smell of the fish oil, improve the intake willingness of dogs and cats on the fish oil, effectively improve the luster and smoothness of hairs of the dogs and cats, and play a role in beautifying hairs.

The invention provides a vehicle operation strategy generation method based on a vehicle-ground integrated electromechanical system. The method can be applied to the technical field of rail vehicle operation and management. The vehicle operation strategy generation method comprises the following steps: responding to received operation state information of a target vehicle on a target path, state information of passengers in each compartment in the target vehicle and first operation environment information in a sight distance range; acquiring second operation environment information in the sight distance range when the target vehicle is located at the target position; determining an operation environment fluctuation type according to the second operation environment information and the first operation environment information; and on the basis of the running environment fluctuation type and the passenger state information in each compartment, a vehicle running strategy is generated by adjusting the running state information of each component, so that the target vehicle can stably run on a path between the target position and the current position under the condition that the passenger requirements are met.

The application provides a semiconductor device manufacturing method and a semiconductor device. The method comprises the following steps: providing a semiconductor substrate, a first trench is formed in the semiconductor substrate, a field oxide layer is formed on the side wall and the bottom wall of the first trench, so that a space outside the surface of the field oxide layer in the first trench forms a second trench, and the first conductive material is filled in the second trench. The field oxide layer of the side wall of the first trench is removed to a first set depth to form a third trench. A barrier layer is formed on the bottom wall of the third trench, a gate oxide layer is formed on the side wall of the third trench, at least two separated gates are formed in the space outside the surface of the gate oxide layer in the third trench, the barrier layer is removed, and the gap between the separated gates is filled. Then, the space between the bottom of the separated gate and the upper surface of the field oxide layer forms an air layer. Due to the introduction of the air layer, the gate leakage capacitance and the gate-source capacitance are reduced, the switching loss and the conduction loss are reduced, and the opening rate of the device is improved.

The application discloses a kind of low-illumination condition fast moving target multi-view cooperative tracking method and device, the method includes low-illumination condition fast moving target multi-view tracking algorithm and unmanned aerial vehicle cluster cooperative tracking algorithm;Multi-source environment perception is carried out using millimeter wave radar sensor, RGB vision sensor and infrared vision sensor carried by unmanned aerial vehicle;Multi-source perception data is normalized in time and space dimensions by three space-time heterogeneous alignment algorithm;Multi-source fusion perception data after space-time normalization is input into multi-source fusion perception feature calculation network to calculate the world coordinates of the tracked target under the current frame;The calculation result is compared with the calculation result of last frame, to judge the change of the position and moving direction of the tracked target in world coordinate system;The world coordinate information of the tracked target is shared through cluster communication, and the trajectory of unmanned aerial vehicle cluster is formulated and updated using multi-agent reinforcement learning, to realize unmanned aerial vehicle cluster cooperative tracking.

The present invention provides a method for extracting low-smelling and bitter codfish bone collagen peptide by using a composite enzyme, comprising the following steps: (1) taking fresh codfish bones, adding deionized water and lipase to carry out enzymatic hydrolysis reaction; (2) washing the fish bones after the reaction, freezing, thawing, repeating the process 2-3 times, and then crushing them into particles with a particle size of ≤5 mm, filtering and drying; (3) adding the composite enzyme into deionized water, adjusting the pH to 6-7 with citric acid, adding the crushed fish bone powder, and carrying out enzymatic hydrolysis reaction in a vacuum state for 6-8 hours; (4) after inactivating the enzyme in a boiling water bath, separating the supernatant and the precipitate from the obtained enzymatic hydrolyzate, and concentrating the supernatant in a vacuum and drying to obtain the codfish bone collagen peptide. The cod bone collagen peptide extracted by the method disclosed in the present invention has a high content of active peptides, with total nitrogen at 13.7-14.2%, oligopeptides at 78.3-79.7%, and protein hydrolysates with a relative molecular mass of less than 1000u accounting for 97.6-98.9%. The extracted cod bone collagen peptide has no obvious bitterness and fishy smell, which is beneficial to improving the application range of the cod bone collagen peptide.

The present application relates to a kind of tetrahydroxy bile acid pharmaceutical composition and its preparation method and application.The present application provides a kind of pharmaceutical composition, it includes 100 weight parts of compound shown in formula I and 300~900 weight parts of saccharide.The present application also provides the use of the pharmaceutical composition in the preparation of drug for treating PFIC.The pharmaceutical composition of the present application can improve dose accuracy, solve the problem of poor taste of bile acid compound, easy to swallow, can improve the compliance of children drug.The pharmaceutical composition of the present application has good stability, easy to transport and store.In addition, the pharmaceutical composition of the present application realizes the stability of treatment effect under different doses, easy to accurately dose children according to body weight, while having good bioavailability, can realize good absorption and utilization in vivo.

The invention relates to a fragment migration method for database middleware and computer equipment. In response to a migration event sent by the management and control node, entering an emigration persistence state, modifying and storing the state of a target fragment of an emigration node as an emigration state; if the emigration state is successfully stored, the state machine enters an emigration state and sends an emigration request to an emigration node, enters an immigration persistence state under the condition that an emigration success event is received, and modifies and stores the state of the target fragment as an immigration state; and if the immigration state is successfully stored, the state machine enters the immigration state, sends a first immigration request to the immigration node, enters a migration ending persistence state under the condition that a first immigration success event is received, and modifies and stores the state of the target fragment as a migration ending state. According to the embodiment of the invention, through the cooperative work of the state machine, the immigration node and the immigration node, the migration of the fragments between the service nodes is completed, the delay of the high-load service node is reduced, and the data access efficiency is improved.

The invention discloses a compressor low-noise operation rotating speed calibration method and device and a storage medium, and relates to the technical field of compressor control, the compressor low-noise operation rotating speed calibration method comprises the steps that rotating speed acceleration noise associated data is acquired, and a to-be-evaded rotating speed threshold value is determined according to the rotating speed acceleration noise associated data; whether the whole vehicle meets a static condition or a dynamic driving condition or not is determined, the vehicle meeting the static condition is in a static state, and the vehicle meeting the dynamic driving condition is in a driving state; when the whole vehicle meets the static condition, compressor noise data and compressor vibration data are determined; when the whole vehicle meets the dynamic driving condition, environment noise data and environment vibration data are determined; and according to the compressor noise data, the compressor vibration data, the environment noise data, the environment vibration data and the to-be-evaded rotating speed threshold value, data matching is conducted, and the target rotating speed of the compressor is obtained. The effect of accurately calibrating the low-noise operation rotating speed of the compressor is achieved.

The application belongs to the technical field of communication, and discloses an AI mobile phone communication man-machine coupling intelligent switching method based on personalized sound replication, which comprises the following steps: S1, receiving a user's voice sample at an AI mobile phone terminal, extracting multi-dimensional voiceprint parameters, and constructing a personalized sound replication model; binding the personalized sound replication model with the user's communication identity identifier to form a communication sound configuration file; S2, when a communication task is triggered, the application scene of the communication task is discriminated based on the initiation mode and target object of the communication task, and a corresponding scene-based communication execution strategy is generated according to the discrimination result; S3, in the process of communication, the speech and semantics of the opposite end are identified in real time, the scene-based communication execution strategy is combined, and the intention of the opposite end and the service response state are jointly determined; the stability of the man-machine coupling communication process and the user experience are improved.

The invention discloses an equipment cascading failure risk assessment method and system based on a distribution network twinborn model. The method comprises the following steps of: respectively generating an N-1-k anticipated accident set and an N-1-1 anticipated accident set when different initial anticipated shutdown elements fail according to a failure state of each equipment element and a topological relation among the equipment elements in a distribution network, wherein the N-1-k anticipated accident set and the N-1-I-1 anticipated accident set are generated when the initial anticipated shutdown elements fail; the method comprises the following steps: extracting a topology connection mode of a distribution network topology judgment initial anticipated shutdown element in a distribution network, and generating an N-1-k anticipated accident set or an N-1-1 anticipated accident set based on the topology connection mode; calculating the failure probability of each equipment element in the distribution network, and calculating the low-voltage risk of all nodes and the overload risk of all lines in the anticipated accident set for each anticipated accident set, so as to generate the distribution network cascading failure risk for each anticipated accident set; and sorting the importance of different anticipated accident sets based on the distribution network cascading failure risk of each anticipated accident set.

The application relates to a low-perception automatic environment lighting adjustment method and system, comprising the following steps: comparing a specific feedback signal collected with a trigger condition set and corresponding to a preset lighting strategy set matched with the trigger; on the basis of pressure collection data generated by walking and moving of the old people, combining a system clock and old people's living habit learning record, comprehensively judging the old people's action intention and behavior implementation purpose, and then automatically adjusting a light source lighting switch and lighting parameters, so that the old people do not need to actively intervene in operation in an environment light source adjustment process, the perception of the old people for the environment light adjustment is reduced as much as possible, and accidents such as falling and knocking caused by unsuitable light environment are avoided, and the best environment light source effect is provided in a natural and soft mode.

The invention provides a gray release method and device, a medium and equipment, and relates to the technical field of computers, in the method, when a gray release demand exists, real-time traffic can be firstly sent to a first-class service until the second-class service is on line, then the real-time traffic is sent to the second-class service, and the first-class service is on line, so that the real-time traffic is sent to the first-class service. Compared with the gray release of the gateway level which is only realized in the related technology, the gray release scheme provided by the invention has stronger integrity, one type of service can call the unique one type of sub-service and apply the unique one type of sub-service to call the address and one type of message queue, and similarly, the service efficiency is improved. And the second-class service can call the unique second-class sub-service and apply the unique second-class sub-service to call the address and the second-class message queue, so that the invention provides an integral gray release solution from the gateway layer to the system bottom layer, is higher in adaptability with an internet micro-service system architecture, and is more convenient to use. The success rate of gray release can be improved.

The application discloses a manufacturing method of bamboo shoot fiber nutrition rice, and belongs to the technical field of food processing. The manufacturing method of the bamboo shoot fiber nutrition rice is as follows: bamboo shoot raw materials are sliced, airflow dried, coarsely pulverized and ultra-micro pulverized to prepare fiber fine powder with a particle size of 1-10 microns; the fiber fine powder is mixed with glutinous rice powder or cassava starch, whey protein powder and the like according to a mass ratio of 7.15-88.97:7.85-85.86:0-36.53 to prepare pulp; 0.3g of glutamine transaminase is added for treatment; three-stage heating and extrusion molding are carried out; and after drying and polishing, the nutrition rice is prepared. The application improves the taste of the bamboo shoot raw materials through ultra-micro pulverization, solves the molding problem through enzyme treatment and segmented heating and extrusion, and prepares the nutrition rice with a dietary fiber content of 5-50.5% and a protein content of 5-30.7%, fills the market blank of high-dietary-fiber nutrition rice, and can fully utilize bamboo shoot resources, and has economic and social benefits.

This invention discloses an integrated communication and sensing method for cloud access networks (CRAN) assisted by an active intelligent reflector. In the context of a CRAN, this method effectively reduces mutual interference between communication and sensing systems by leveraging the centralized signal processing advantages of a multi-antenna baseband processing unit (BBU). Simultaneously, it utilizes an active intelligent reflector (Active-IRS) to assist both communication and sensing. A joint design of a multi-antenna remote radio head (RRH) and beamforming at the active intelligent reflector ensures that the reflected signal from the active intelligent reflector is aligned with the target direction. The amplifier integrated into the active intelligent reflector enhances the energy of the reflected signal to counteract the impact of multipath loss on signal strength. This method achieves the desired reflector beam while maintaining the user's signal-to-interference-plus-noise ratio (SNR).

The utility model relates to the technical field of vehicle safety, in particular to an anti-submergence type seat and a vehicle. The anti-diving type seat comprises a seat body, the seat body is provided with a bearing face suitable for a passenger to sit, and the bearing face is downwards and obliquely arranged from the front end of the seat body to the rear end of the seat body; the supporting main body is fixedly connected to the seat body, the supporting main body is provided with a first load-bearing surface, and the first load-bearing surface is obliquely arranged downwards from the front end of the seat body to the rear end of the seat body; and the buffer structure is arranged on the first loaded surface, and the buffer structure is suitable for deforming when being subjected to external impact force. The supporting body is fixedly connected to the seat body, the installation stability of the anti-descending structure is guaranteed, the buffering structure is arranged on the first loaded face of the supporting body, the buffering structure deforms when being subjected to external impact force, the buffering and energy absorbing effects are achieved, and passengers are protected; therefore, stable installation of the anti-diving structure is ensured while anti-diving protection and buffering protection are carried out on passengers in the frontal collision process.

The present disclosure provides a large model request processing system and an electronic device. A large model request processing method includes: a terminal device responding to an information transmission exception message of a target model, determining a transmission exception cause; the terminal device sending a continuation request to the server according to the transmission exception cause, the continuation request including question information and displayed information; the server forming continuation information according to the question information and the displayed information of the continuation request; the server sending the continuation information to the terminal device; and the terminal device displaying the continuation information after the displayed information. The embodiments of the present disclosure can optimize the user experience of a large model.