33results about How to "Automate selection" patented technology

The invention discloses an image segmentation method and device and belongs to the field of image processing. The image segmentation method includes the following steps: establishing a significance model of an image; according to the significance model, obtaining foreground sample points and background sample points in the image; according to the significance model and the foreground sample points and the background sample points, establishing a foreground and background classification model; and according to a preset image segmentation algorithm, segmenting the image, wherein the preset image segmentation algorithm uses the foreground and background classification model and edge information between pixel points to segment the image. Through automatic determination of the foreground and background sample points in combination with the significance model, the foreground and background classification model is established and the foreground and background classification model is used to realize image segmentation. Therefore, problems, which exist in related technologies, that the foreground sample points and the background sample points must be selected roughly manually and the segmentation efficiency is comparatively low when a large quantity of images are segmented are solved so that effects of realizing automation selection of samples and improving the classification precision and segmentation efficiency are achieved.

The invention discloses a video cover extraction method and device based on a video title. The method includes: acquiring a cover set of a video, wherein the cover set contains at least one to-be-selected cover image; inputting the video title of the video into a semantic-feature extraction module, which is obtained through training, to obtain a video title feature; inputting each to-be-selected cover image in the cover set into an image feature extraction module, which is obtained through training, to obtain a cover image feature of each to-be-selected cover image; and calculating a distancebetween the video title feature and the cover image feature of each to-be-selected cover image, and determining a should-be-selected cover image according to a calculation result. By adopting the scheme, automated selection of a video cover with a high relevance degree with the video title can be realized, a matching degree of the video cover and video contents and the title can be increased, anduser experience can be improved.

The invention discloses an intelligent Internet-of-Things blind guide stick. A vibration module and a Bluetooth headset module are arranged in a hand-shaped stick handle. A radar ranging module, a GPS positioning module, an LED illumination module, an image recognition module, a bar code recognition module and a cloud data storage module are arranged in a stick body. A charging interface is installed at the bottom of the stick body. The vibration module comprises a microprocessor, a vibration class memory, a vibration time memory and a motor. The Bluetooth headset module comprises a Bluetooth chip, an audio processing chip and an audio emitting module. A voice recognition module and an image recognition module are arranged in the vibration module. Through the intelligent Internet-of-Things blind guide stick, exchange between road information and a user, recognition of the voice of the user, making of a walking route and recognition of blind sidewalk and bus information are realized; whether an obstacle exists in front or not is sensed through ultrasonic waves, meanwhile, a prompter emits sounds to notify the blind, the stick can adapt to people with different heights, and a lamp can caution surrounding pedestrians and vehicles for avoidance; the stick is simple in structure and convenient to use.

The invention relates to an end-effector device and automated selective thinning system. The system includes vision acquisition hardware, kinematic targeting and heuristic programming, a robotic arm, and a pomologically designed end-effector. The system is utilized to improve efficiency for the fruit-thinning process in a tree orchard, such as peach thinning By automating the mechanical process of fruit thinning, selective fruit-thinners can eliminate manual labor inputs and further enhance favorable blossom removal. Automation used in conjunction with a heuristic approach provides improvements to the system. The system may also be configured as a robotic arm or as a handheld system by including a battery and switching microcontroller with handle or wrist straps. Handheld thinning devices that are mechanical in nature may also be part of the system.

A test chip socket comprises a contact block fixed to a tester substrate, a tray mounted to the contact block, the tray having a convex part for positioning test target chips to a plurality of mount positions, and a plurality of probes each of which is held by the contact block and contacts the tester substrate and the test target chip, wherein each probe contacts with a terminal of each test target chip mounted to the mounting position when the tray mounting the plurality of the test target chip is fixed to the contact block.

The present invention relates to an intelligent disposition and process monitoring system based on cloud management platform. The system comprises an order processing module, an information module, an order analysis module, an intelligent disposition module, a failure information route processing module and a monitoring module. The order processing module is configured to submit an order for users and perform ordering checking; the information module is configured to receive and process the order after the checking passes; the order analysis module is configured to analyze the order and obtain the disposition task and the disposition strategy in the order; the intelligent disposition module is configured to perform intelligent disposition; the failure information route processing module is configured to perform failure reason analysis and feed failure reason analysis back to the intelligent disposition module; and the monitoring module is configured to associate with the order disposition log in the order with successful disposition and monitor and inquire the order disposition log. The intelligent disposition and process monitoring system and method based on a cloud management platform are able to greatly reduce the manpower and material resources invested in the process of disposition, management and upgrading maintenance and decrease the operation cost.

The present invention generally relates to a set of early developmental reference data or “lineage scorecard” for stem cells, and methods, systems and kits to generate a lineage scorecard for predicting the functionality and suitability of stem cell lines. In some aspects, methods for generating a scorecard comprises measuring the gene expression of a plurality of early developmental genes, such as pluripotent, early ectoderm, early mesoderm and early endoderm genes to predict the pluripotency and differentiation potential of the stem cell line and its functionality and/or suitability for a desired use. In some embodiments, a reference scorecard can be compared with the test stem cell line scorecard to accurately predict the utility and/or identify specific characteristics of the stem cell line, e.g., to determine its suitability for downstream applications, e.g., therapeutic use, drug screening, toxicity assays, differentiation into a desired cell lineage, etc.

The invention discloses a GAN-based low-cost adversarial network attack sample generation method, comprising the steps: 1, encoding attributes and labels of samples in a sample set; 2, calculating aninformation gain according to the attribute of the coded sample; 3, training an adversarial sample generation model; and 4, generating an adversarial sample for attacks. Based on the basic thought ofGAN, the GAN-based low-cost adversarial network attack sample generation method realizes automatic selection of disturbance characteristics by calculating attributes, labels and information gain training models of samples, and can generate efficient adversarial network attack samples at the minimum cost.

The invention discloses an optimization design method for a Whiffletree support point position of primary mirror and sub-mirror models of a telescope. The method comprises the following steps that (1)a sub-mirror model is simplified; (2) the quantity of Whiffletree support points is determined, and parametric representation is carried out on each support point; (3) parametric models of the sub-mirror and support points are constructed, initialization is carried out, constraint and loading are applied to perform first statics analysis, and all nodal displacement on the upper surface of the sub-mirror is obtained; (4) root-mean-square values RMSs of all the nodal displacement are calculated; (5) a position parameter of each support point is taken as an optimization variable, the minimum RMSis taken as an optimization objective, so that the optimization is carried out, and the optimized Whiffletree support point position is obtained. According to the method, whole process steps and parameters used by each key step of the optimization method are improved, reasonable support point position parameters can be obtained, and the surface precision of the upper surface of the sub-mirror satisfying design requirements is obtained.

The invention discloses a double drive planetary gear reducer and a hydraulic clutch device. The reducer and the device comprise a planetary reducer base and a middle gear, a connection flange and a clutch base. The middle gear, connection flange and the clutch base are sequentially arranged in the axial direction of the planetary reducer base. A clutch I and a clutch II are placed at two sides of the planetary reducer base. The axial directions of the clutch I and clutch II are in parallel with the axial direction of the planetary reducer base. The middle gear is connected with the planetary reducer base through a ball bearing, and is connected with the connection flange through a roller bearing. The double drive planetary gear reducer and the hydraulic clutch device achieve the coaxial disconnection and connection with reliable connection status, and are flexibly and agilely detachable. The switching between the clutch I and the clutch II is flexible and convenient. The structures of both the clutch I and clutch II are compact. The automatic selection between the single-input-driving-and-switching and double-input-driving-and-switching is achieved. The piston and the clutches are arranged on a fixed shell body. The oil line setting and control are convenient. The linkage control is easily achievable. The input shaft structure is simple. There is no necessity to arrange the bearing structures to sustain the axial direction force.

The invention provides a feature combination method and device for content recommendation, equipment and a storage medium. The method comprises the following steps: constructing a feature combinationmode set comprising at least one feature combination mode; respectively determining the effectiveness of each feature combination mode; based on the effectiveness of each feature combination mode, screening from the feature combination mode set to obtain a first target number of feature combination modes as first candidate combination modes; generating at least one second candidate combination mode based on the first candidate combination mode and the features; selecting a feature combination mode meeting a screening condition from the at least one second candidate combination mode as a targetfeature combination mode; based on the target feature combination mode, performing feature combination on the features to obtain target combination features; by means of the method and device, automatic selection of the feature combination modes can be achieved, and the effect of intelligent content recommendation based on the combination features obtained through the selected feature combinationmodes is improved.

Embodiments of the invention provide an interference-free point selection method and device for an optical switch. The method comprises the following steps of dividing channel points of the optical switch into a plurality of groups; aiming at any group, establishing a grid point matrix of the group, wherein the grid point matrix is composed of a plurality of grid points; if a relation between anyone of the grid points in the group and any one of the channel points meets a first preset rule, adding the channel point into a channel point set of the grid point; based on the size of the channel point set of each grid point in the group, selecting the to-be-matched grid points of the group; and aiming at all the groups, if a relation between the to-be-matched grid points of every two groups meets a second preset condition, taking the to-be-matched grid points of each group as interference-free points. According to the method and the device provided by the embodiments of the invention, automatic selection of the interference-free points of the optical switch is realized; the selection rate and the accuracy rate of the interference-free points are improved; and the condition is providedfor the optical switch to realize efficient and accurate interference-free channel switching.

The invention discloses a cherry picking device, which comprises a conveyor belt and a picker located directly above the conveyor belt; a first baffle plate and a second baffle plate are arranged on both sides of the conveyor belt, and one side of the first baffle plate and the One side of the second baffle is respectively provided with a first fixed seat and a second fixed seat, the first fixed seat and the second fixed seat are provided with a funnel groove, and the selector is located at the outlet just below the funnel groove, and the selector The first rotating shaft and the second rotating shaft are respectively provided on both sides of the motor, the first rotating shaft is connected with the output shaft of the motor, and the second rotating shaft is rotatably arranged on the second fixing seat; The selector is cylindrical, and a plurality of first selection holes are evenly distributed on the side of the selector. Cherries of different sizes can be sorted and selected to realize the automatic selection of cherries. The selection efficiency is high, the error is small, time-saving and labor-saving, and it is extremely convenient to use.

The invention discloses a mirror image source management method, a resource downloading method and computing equipment. The mirror image source management method is executed at a server, and a recommended mirror image source set can be generated according to the working state of each mirror image source. When a user has a resource downloading requirement, the user can download; the server sends the recommended mirror image source set to terminal equipment where the user is located; the terminal equipment does not need manual participation in selection; According to the method, a plurality of mirror image sources which are optimal for the current environment of the user are selected from the recommended mirror image source set to serve as the target mirror image sources, resources are requested from all the target mirror image sources respectively, and therefore automatic selection of the mirror image sources and distributed downloading of the resources are achieved, downloading efficiency is improved, and the user can quickly and efficiently complete updating of the system and the application.

The invention relates to a manufacturing method of an electrochemical sensor, and provides a manufacturing method of an electrochemical sensor, which is rapid, high in quality, competitive in price, simple and convenient in process control and suitable for large-scale production. According to the manufacturing method, a PCB or other plastic substrate or other ceramic substrate or other metal ceramic substrate is adopted, and mixed electroplating, thick film screen printing and a micro-manufacturing process based on silicon wafer processing are used for manufacturing the electrochemical sensor.Different metal material thickness/precision requirements and a relative adding sequence on the sensor can determine process selection.

The invention provides an automatic berthing method, device and system for an intelligent ship, and the method comprises the steps: obtaining a berth list capable of being berthed of a pre-berthing wharf, selecting a first berth capable of being berthed closest to the ship in the berth list capable of being berthed, and transmitting a berthing request to the pre-berthing wharf; after parking allowing information sent by the pre-parking wharf is received, environment information and weather information of an area where the ship is located are obtained, position information of the ship and the first berth capable of being parked is obtained, and a berth path is generated; and according to the berth path, the current driving direction of the ship is adjusted in time until the ship is berthed at the first berth capable of being berthed. Compared with the prior art, the technical scheme provided by the invention can reduce the investment of manpower and material resources, and improves the ship berthing efficiency at the same time.

The invention relates to a brake driving mechanism operation mode selection system which comprises a brake driving mechanism comprising an autombile brake structure and an automobile driving structure. The automobile driving structure comprises a flywheel, a pressing plate, a clutch plate, a speed changer, a gear shifting control mechanism, a universal joint, a hub, explosion-proof video recording equipment, a dust extraction structure, a dust storage box body, content adjusting equipment, signal enhancement equipment, an instrument box body and contrast ratio improving equipment. The automobile brake structure comprises a brake main pump, a brake cylinder, a vacuum booster, a brake pedal assembly, a brake disc, a brake drum, a brake pad and a brake oil pipe. The invention further relates to a brake driving mechanism operation mode selection method. The brake driving mechanism operation mode selection system and method are compact in structure and convenient to use. The cleaning mode of the dust in the tire can be automatically selected based on the high-precision analysis result of the number of the dust particles between the hub and the tire, so that excessive dust collection in the tire is avoided.

The invention belongs to the field of biotechnologies, and discloses a multi-functional intelligent portable microorganism collecting system. The multi-functional intelligent portable microorganism collecting system comprises a solar energy power supply module, an image collecting module, a temperature detecting module, a central control module, a drying module, a culturing module, a detecting module, and a display module. The multi-functional intelligent portable microorganism collecting system is capable of getting a continuous solar energy through the solar energy power supply module, and guaranteeing the lasting work of the collecting system, energy-saving, economical and environment-friendly; is conveniently powered at any time in a carrying process; and meanwhile is capable of rapidly and effectively detecting the content of microorganisms (especially bacteria, fungi or viruses and the like) in a sample through the detecting module. The multi-functional intelligent portable microorganism collecting system is simple and easy to operate, strong in specificity, high in sensitivity, and has the application prospect in the trace microorganism detection field.