176results about How to "For subsequent analysis" patented technology

The invention discloses a video behavior timeline detection method. The method comprises the following steps: performing modeling based on a deep learning and time structure, detecting a video behavior timeline in combination with coarse granularity detection and fine granularity detection, and extracting temporal and spatial characteristics of a video by using a dual-flow model on the basis of anexisting model SSN; modeling a time structure of a behavior, and dividing a single behavior into three stages; then providing a new characteristic pyramid capable of effectively extracting time boundary information of a video behavior; and finally, combining the coarse granularity detection and the fine granularity detection to make a detection result more precise. The video behavior timeline detection method is high in detection precision, and the detection precision is higher than the detection precision of all of the current existing disclosed methods; the video behavior timeline detectionmethod is wide in application range, is applicable to detection of video clips in which people are interested in an intelligent monitoring system or a human-machine supervision system, is favorable for subsequent analysis and processing, and has an important application value.

The invention discloses a method for separating high-purity circulating tumor cells from blood, comprising the following steps of: A. loading single chain polynucleotides on the surface of a micro channel of a micro fluidic chip, and enabling the surface of the micro channel to load an antibody by performing specific binding on polynucleotides and mark single chain polynucleotide in an antibody-mark single chain polynucleotide compound; B. contacting a blood sample with the antibody loaded on the surface of the micro channel of the micro fluidic chip, and separating the circulating tumor cells from blood by the combination of the antibody with surface specific antigens of the circulating tumor cells in the blood; and C. implementing release of the antibody and the captured circulating tumor cells from the surface of the micro channel by cutting a polynucleotide chain, thereby separating to obtain the high purity circulating tumor cells. The method provided by the invention can separate the high purity circulating tumor cells. If the released cells further pass through the loaded chip aiming at hemameba, the circulating tumor cells with sky-high purity can be obtained.

The invention provides a storage box recognition and positioning method based on contour features. The method comprises the following steps: (1) collecting a color image and a depth image of box goods; (2) performing foreground segmentation on the color image and the depth image, mapping depth blocks to the color image, marking color region blocks falling into the same depth block, and combining the color region blocks to serve as a final box cargo foreground image; (3) performing linear edge detection on the foreground image of the box cargo, and connecting edges by taking a rectangle as an outline target to obtain a candidate target closed outline; (4) selecting the target meeting the predetermined box body rectangular characteristics from the candidate target closed contours to obtain abox body identification result. The method performs identification and positioning of the box goods based on the depth image and the color image, improves the calculation efficiency, has the advantages of low cost and high real-time performance, and is suitable for identification and positioning of the box goods in complex environments such as factory warehouses.

The invention provides a road condition identifying method for lower limb prosthesis walking and relates to lower limb prostheses. The method comprises the steps of using a front sole pressure sensor, a rear heel pressure sensor and an inertial measurement unit module to collect residual end movement signal sample data of a tested person, experimental road conditions including flat ground, upward slope, downward slope, upward stairs and downward stairs; selecting the starting point of road condition identifying time; at the moment starting to use the inertial measurement unit to collect the residual end movement signal sample data of the tested person; determining the number for practical use of the collected residual end movement signal sample data; processing the collected and calculated residual end movement signal sample data; calculating the characteristic valve of the collected residual end movement signal sample data; training with an LVQ-ADABOOST classifier; realizing online identification of the road conditions in lower limb prosthesis walking. The problem that the offline road condition identifying method for lower limb prosthesis walking cannot guarantee the instantaneity of road condition identification is solved.

The invention relates to a total-stroke and total-space transient electromagnetic device and measuring method. The device comprises a transmitting unit and a multi-channel receiving unit, wherein the transmitting unit and the multi-channel receiving unit are respectively connected with a signal processing unit, the signal processing unit is connected with a main engine through a communication unit, relevant parameters are set through operation software of the main engine, the receiving unit and the transmitting unit jointly achieve field source excitation and complete collection, recording and the like of an induction signal and an excitation source signal, the induction signal and the excitation source signal are transmitted to the main engine through an RS485, and the display, storage, management and the like of the data are achieved. According to the total-stroke and total-space transient electromagnetic device and measuring method, the self-induction and mutual induction influences can be effectively reduced, the signal to noise ratio and the sensitivity are improved, receiving coils can be combined into different types according to the detection requirement, the different detection effects can be obtained, and the judgment on the direction of an abnormal body can be achieved under the total-space situation. The number of detection blind zones can be reduced, the superficial zone detection effect can be improved, an ARM quad-core processer hardware platform is adopted, a special embedded type operation system is configured, and real-time imaging, comparison, judgment and subsequent analysis can be conveniently carried out on the measured data.

The invention discloses an overall performance testing rack of a numerical control ultrasonic cutting sound main shaft. The overall performance testing rack comprises a base. Perpendicular linear guide rails, an artificial load device and a radial movement base are installed on the base. The perpendicular linear guide rails are in sliding fit with the ultrasonic cutting sound main shaft. The artificial load device is provided with two force application rods. The two force application rods are located at the axial position and the radial position of a cutting tool nose of the ultrasonic cutting sound main shaft respectively and used for applying artificial loads to the cutting tool nose. The radial movement base is provided with a testing probe which can move towards the ultrasonic cutting sound main shaft in a radial mode and rotate. According to the overall performance testing rack of the numerical control ultrasonic cutting sound main shaft, the testing site and various costs can be reduced, the measurement on different parameters under the same installation condition is provided, the testing errors under different installation conditions are reduced, the testing effect and the testing efficiency on the performance of the ultrasonic cutting sound main shaft are improved, and test bases are provided for the further perfection and the further improvement of the performance of the ultrasonic cutting sound main shaft.

The invention provides a monitoring system for a water injection state of each layer of a multi-layer water injection well. The system comprises a ground control main machine and a sleeve, wherein a carrier steel pipe cable, cable passing packers and multiple underground water distribution devices are arranged in an inner cavity of the sleeve; the multiple underground water distribution devices are connected with the carrier steel pipe cable and an oil pipe in the sleeve; each underground water distribution device is corresponding to one distribution stratum, and is used for acquiring injection state data of the corresponding distribution stratum in real time; the cable passing packers are connected with the oil pipe, and are packed at the two ends of each underground water distribution device, so that a sealed water injection cavity is formed at each place where an underground water distribution device is situated; the carrier steel pipe cable is used for transmitting the injection state data of each distribution stratum acquired by each underground water distribution device to the ground control main machine in real time; and the ground control main machine is used for storing the injection state data of each distribution stratum. The system can monitor the injection state data of the water injection well in real time for a long time, and prevents the scene monitoring.

The present invention discloses a medical insurance data-based clinical pathway analysis method. The method comprises a medical insurance data cleaning step and a medical insurance data analyzing step. During the medical insurance data cleaning step, invalid hospitalization data and invalid prescription detail data are removed. Meanwhile, data are clustered based on identical disease entity numbers and identical hospital level numbers, so that a plurality of analysis units are established. During the medical insurance data analyzing step, medical insurance analysis items in the analysis unitsare subjected to empirical range analysis. The empirical range analysis is in the form of grouped empirical range analysis or non-grouped empirical range analysis, wherein the sample most aggregated distribution range of medical insurance analysis items is calculated. When the frequency sum of the most aggregated distribution range is larger than a preset frequency, the empirical range of the medical insurance analysis items is obtained. According to the invention, medical insurance data are cleaned and analyzed, and then the clinical pathway analysis is conducted based on medical insurance big data. The accuracy and the scientificity of the clinical pathway analysis are improved.

The invention relates to a micro-plastic extraction device based on the electrostatic technology. The micro-plastic extraction device based on the electrostatic technology comprises a machine box, a feeding device, a high-voltage composite electrode, a roller electrode and hoppers, wherein the feeding device, the high-voltage composite electrode, the roller electrode and the hoppers are arranged in the machine box. One end of the feeding device is connected with the box wall of the machine box. The other end of the feeding device is located above the roller electrode. The high-voltage composite electrode is arranged above the right side of the roller electrode. An electrostatic field is formed between the high-voltage composite electrode and the roller electrode, so that air is ionized, and a great number of electrons are generated. The hoppers are located below the roller electrode. According to the micro-plastic extraction device based on the electrostatic technology, effective separation and extraction are conducted on micro-plastics in silt through the corona electrostatic technology, the workload of manual extraction and separation is greatly lightened, and the separation efficiency is improved. The micro-plastic extraction device based on the electrostatic technology can be widely applied to the technical field of environment monitoring and the technical field of separating micro-plastics from water body sediments.

The instrument includes body case and circuit board. Following parts are setup on case: high light character screen, tens minute place screen, minute place screen etc. remote control receiving header, manual reset, stop and start push buttons, preset push buttons, online data communication port, volume control knob, speaker and power cable etc. remote controller and manual operation can preset, start, stop and reset the instrument. Count down and clocking forward in case of overtime are displayed on digital large screen. Functions of voice hint and voice report are provided. Time data is kept in case of power failure. Data can be transferred to computer through online data port. The invented instrument is suitable to controlling time limit, such as interview, lecture etc.

The invention discloses an upper limb exoskeleton mechanical arm and a control method of rehabilitation training, and the rehabilitation training is guided by medical staff and self-adaptive to patient limbs. According to the invention, the movement trajectory of rehabilitation training is recorded by a teaching mode, patient exercise intensity is identified by a deep learning framework, the identified results are embedded into an overall rehabilitation training control strategy based on deep reinforcement learning, and the adjusted joint movement rate signals are output to the mechanical arm to complete the rehabilitation training. On the premise of preventing secondary injuries of patients from the rehabilitation training, the method can eliminate the influence of hysteretic nature of movement signals and individual differences of bioelectric signals, improve the identification accuracy of exercise intensity and broaden the application area of rehabilitation movement trajectory by the vector fusion of movement signals and biological signals, so as to achieve advantages of reducing the burden of medical staff and optimizing the rehabilitation cycle of patients, thus improving the safety and applicability of the rehabilitation training.

The invention discloses a self-adaptive denoising method for partial discharge signals on the basis of spectral kurtosis and S transformation. The self-adaptive denoising method comprises the following steps of: through a discrete signal rapid spectral-kurtosis diagram and S transformation time-frequency spectrum calculation, rapidly acquiring time-frequency characteristics of the partial discharge signals; by combination of the time-frequency characteristics of the partial discharge signals, optimizing value ranges of all parameters in a partial-discharge-signal matching atom dictionary in a self-adaptive manner; then finding atoms with maximum inner product with noised partial discharge signals as optimum matching atoms from the optimized results of the partial-discharge-signal matching atom dictionary by an MP (Matching Pursuit) algorithm; finally, on the basis of the optimum matching atoms obtained by all iterations, carrying out sparse representation on the original partial discharge pulse signals so as to achieve the purpose of denoising the partial discharge signals. The self-adaptive denoising method disclosed by the invention has the advantages that the accuracy of the denoising result is high, no waveform distortion is caused, no limitation on the number of measured local discharge pulses and the positions of the pulses is formed, and the application is wide.

The invention relates to a wearable lung sound detection device, which comprises an acoustic-electric conversion sensor and a soft plastic suction cup. The front of the sensor extends out of the shell and enters the inner cavity of the soft plastic suction cup installed on the shell. The circuit board of the lung sound collection is equipped with a microprocessor, a lung sound signal processing module, an A/D sampling module and a power supply module. The acoustic-electric conversion sensor is used to transmit the collected lung sound signal to the lung sound signal processing module, and the lung sound signal processing module is used to amplify and filter the collected lung sound signal to the A/D sampling module, and the A/D sampling module The preprocessed lung sound signal is A/D sampled and the sampled digital lung sound signal is saved as an audio file, and the microprocessor communicates with the intelligent terminal through a wireless transmission module. It can accurately collect digitized lung sounds, and realizes a wearable lung sound collection device.

The embodiment of the invention discloses a film watching information recording method and device for a video-on-demand cinema. The method comprises the steps that compartment players receive watching films sent by a local server and play the watching films; and after the playing is finished, the compartment players send film watching information of the watching films to the local server, so the local server sends the film watching information to a cloud server for recording. The method also comprises the steps that the local server receives triggering operation of selecting the films by users and sends the corresponding watching films to the compartment players corresponding to the users according to the triggering operation; and the local server receives the film watching information sent by the compartment players and sends the film watching information and user identifiers of the users to the cloud server for associated recording. According to the embodiment of the invention, the film watching information is sent to the cloud server for recording by the compartment players, the film watching data of the users corresponding to the compartments can be recorded conveniently, thereby facilitating follow-up analysis of film watching habits of the users, and the better film watching service is provided for the users.

The invention discloses a biological sequence analysis platform which comprises a computer, an input module, an analysis module and an output module, wherein the computer comprises a user interface; the input module is used for manually inputting a biological sequence to be analyzed on the user interface, or uploading a readable-format file recording the biological sequence; the analysis module is used for analyzing the biological sequence obtained by the input module to obtain an analysis result; and the output module is used for displaying the analysis result obtained by the analysis module on the user interface, and/or exporting and storing the analysis result in a pre-defined external file outside the biological sequence analysis platform. The biological sequence analysis platform disclosed by the invention has a simple interface, realizes one-stop analysis, and remarkably saves labor and material resources; meanwhile, by adding a batch processing module, the result visualization and storage are enhanced to facilitate the subsequent analysis and view; and in the analysis process, the powerful search algorithm is improved, and the corresponding analysis time is shortened.

The invention provides a method for predicting a tight oil fracturing range based on a DL model of physical constraints, and belongs to the technical field of tight oil fracturing. The method comprises the following steps: firstly, providing a reasonable physical hypothesis for a studied problem and setting related parameters, then establishing a two-dimensional axisymmetric numerical calculation model, dividing a network and setting time steps, establishing a deep learning model, setting a structure and other parameters of a deep neural network model, and obtaining a prediction result by using a test set; and evaluating a prediction result by using an L2 norm and a decision coefficient R2 as evaluation indexes, comparing the prediction precision of the neural network under the condition of considering the physical constraint and the prediction precision of the neural network under the condition of not considering the physical constraint, and finally changing the training data size on the basis of the neural network added with the physical constraint and measuring the prediction capability of the neural network. The method can be applied to rapid prediction of flow field distribution of two-drive model partitions, is high in prediction accuracy and adaptability, is high in calculation speed, and can well solve the problem of unknown tight oil flow field distribution and the problem of small sample prediction.

The invention provides a preprocessing method for automatic water quality monitoring data. The preprocessing method comprises the following steps: S1, counting and marking missing values in an original data set; S2, performing time sequence decomposition on the data in the original data set to obtain a residual term; S3, identifying an abnormal value in the residual item, and deleting and/or correcting the abnormal value; and S4, interpolating missing values and deleted abnormal values in the original data set by adopting a linear interpolation method to obtain a processed complete data set. Compared with the prior art, the preprocessing method has the advantages that abnormal values and missing values in the automatic water quality monitoring value set are effectively processed, and a complete and high-reliability data set can be obtained, so that convenience is provided for subsequent analysis and application of automatic water quality monitoring data.

The invention provides an automated testing method for application performance. The automated testing method is based on an Android system mobile terminal. In the testing method disclosed by the invention, a script program is combined with an adb tool to automatically obtain APP performance, the APP performance can be recorded under a situation of no-man monitoring so as to be favorable for subsequent analysis; and a script program is installed in a computer, resources in the mobile terminal are completely released during testing, and only an APP to be tested operates, the resources of the mobile terminal are not occupied, and a testing result is accurate.

The invention discloses a data storage and backtracking platform for N3 electric sweepers. The data storage and backtracking platform is composed of an online end and an offline end. The online end isused for carrying out data online acquisition and storage on the road garbage information, collecting the current upload gear operation of the sweeper, providing data for the upload gear selection module and providing the optimal upload gear selection under the current garbage scene for the online vehicle speed optimization control module so as to realize decoupling of the upload operation and vehicle speed control; the off-line end performs uploading gear selection algorithm training by utilizing the data collected by the on-line end to obtain relevant parameters in an uploading gear selection algorithm and send the relevant parameters to the on-line end uploading gear selection module for iteratively updating an uploading gear control algorithm of the on-line end; the neural network parameters from the online vehicle speed optimization control modules of the plurality of sweepers are fused and then transmitted to each sweeper; and the on-line end and the off-line end realize data transmission through a control information updating system.