36results about How to "Method smart" patented technology

The invention discloses a visible-light and infrared heterologous image recognition method based on mixed features. The method comprises the following steps: utilizing a self-adaptive canny operator, which is based on an improved Otsu method, to extract edge information, and acquiring edge images of a visible-light image and an infrared image; carrying out multi-scale FAST feature detection on the edge images of the visible-light image and the infrared image, and obtaining main directions of feature points according to a gray centroid method; respectively carrying out ORB feature description on feature point point sets, adopting Hamming distances to measure similarity between the feature points according to a nearest secondary-near principle to carry out rough matching to respectively obtain rough-matching feature point point sets of the edge images; and utilizing an improved RANSAC algorithm to eliminate mismatch point pairs therein, and showing a result in the visible-light image and the infrared image. According to the method, matching of the visible-light image to the infrared image can be effectively realized, and the algorithm efficiency and the matching accuracy can be improved.

A Risk control platform construction method is disclosed. The method includes obtaining actual business requirements; Process configuration and decision configuration of the rule engine according tothe preset function according to the actual business requirements; Acquiring test data, testing the rule engine using the test data based on the result of the process configuration and the result of the decision configuration, and obtaining a test result; If the test result is in conformity with the expected goal, the rule engine is deployed on line; The rule engine is jointly tuned, and the ruleengine is put on line after the joint tuning is successful. The present application also provides a risk control platform construction system that implements the above method. The present applicationsolves the technical problems in the related art that the credit approval risk control is not targeted and the reliability is poor due to the fact that the risk control rule components cannot be freely combined and configured.

The embodiment of the invention discloses a method and device for controlling a parking place lock. The method includes the steps that user information is obtained through a user information obtaining device, and the user information is matched with preset parking place information, wherein the preset parking place information comprises a mapping relation of the user information and parking place position information; if the user information is matched with the preset parking place information, a parking place unlocking command corresponding to the position information is obtained through a parking place information storage device; the unlocking command is received through a parking place detection terminal, and the parking place lock is opened according to the unlocking command. The embodiment of the invention discloses the method and device for controlling the parking place lock. The unlocking process is simplified, labor cost is reduced, and a more intelligent and efficient method is provided for a user.

The present invention discloses an information processing method and an electronic device. The method comprises: acquiring a first parameter; according to the first parameter, determining N functions corresponding to the first parameter in M functions, wherein, M is a positive integer, and N is a positive integer less than or equal to M; and according to N functions, displaying first display content in a display unit of the electrical device, so that a viewer can simultaneously perceive the first display content and the content corresponding to an environment where the electrical device is located. The present invention provides the method for solving the technical problem in the prior art that an electrical device is incapable of automatically changing display content.

The invention discloses an instant video display method and an electronic device and belongs to video field. The method comprises: acquiring a first display parameter input by a user, wherein the first display parameter is used for adjusting a display ratio of an instant video frame; displaying the instant video frame according to the first display parameter; acquiring a second display parameter if a screen is rotated, wherein the second display parameter is used for adjusting the display ratio of the instant video frame after screen is rotated; and displaying the instant video frame according to the second display parameter. The method adjusts the display ratio of the instant video frame according to the first display parameter or the second display parameter so as to increase the display modes and the effects of the instant video frame, satisfy a user requirement, and improve user experience.

The invention provides an intelligent sewing device based on machine vision. The intelligent sewing device comprises a sewing machine control device, a sewing machine, a first moving device, a secondmoving device, an image acquisition device, a light source and a processing device. The invention further provides an intelligent sewing method based on machine vision. A collar sewing system based onmachine vision is introduced, and a more stable system is achieved. According to the device, frequent and complicated collar path planning is omitted, and a simple calibration method is introduced. The method is more intelligent and accurate, a sewing path parallel to a collar can be automatically planned, and sewing cannot be affected after the collar is deformed. An intelligent needle adjustingalgorithm is introduced, the distance between the needles is automatically adjusted, sewing points fall on intersection points of the sewing path, and the sewn beautiful appearance of the collar is ensured. According to the device and the method, and dependence of the system on skilled operators is reduced.

The invention relates to a machine fault detection, classification and grading method based on neural network unified modeling, and belongs to the technical field of machine fault detection methods and artificial intelligence. Firstly, the rotating speed, temperature, vibration and sound of a to-be-tested machine in the operation process are collected, sample annotation is added to sample featuresaccording to a machine fault annotation log and fault grade duration, and a classified and graded fault sample set is generated; samples are randomly extracted from the samples in the fault-free state to form a fault-free sample set, wherein the fault sample set and the fault-free sample set form a complete machine state sample set. The fault types and levels are coded according to a binary mode,and codes of different types and different levels are spliced into a code with unified fault detection, classification and grading. A deep neural network model is established and trained in a unifiedmanner, the model obtained through training has higher diagnosis accuracy and prediction capability, and grading results of various fault types can be diagnosed at the same time under the condition that various fault types coexist.

The invention provides an S-mode selective calling arrangement method. All the targets of the current beam are extracted from an inquiry list to be grouped and arranged, and all the targets of one time of selective calling arrangement are determined by considering the distance difference between the two adjacent targets of each group and the fact that the electromagnetic wave round trip time of the last target of each group is greater than the sum of the duration time of all the previous response pulses and the sum of the inquiry time of all the targets of all the groups is less than the residence time of one wave beam; and the target message transmitting moment of each selected target is calculated and the target message transmitting moment of all the selected targets of one selective calling inquiry period is acquired, and a time table is formed according to the target message transmitting moment of all the selected targets to perform inquiry pulse arrangement to be transmitted to a transmitter. Compared with the methods in the prior art, the inquiry arrangement method is given for multiple targets in one S-mode selective calling period so as to be effective and intelligent; meanwhile, the influence of the factors of the inquiry repetition frequency, the target number and the distance difference between the targets is considered so that the inquiry efficiency can be enhanced in the aspects of time and space.

The invention provides a raw gemstone machining method. The raw gemstone machining method comprises the steps that a plurality of image collecting devices and a host machine are provided, wherein the image collecting devices obtain gemstone image information and transferring the gemstone image information to the host machine, and the host machine judges whether internal jade has been exposed out of the two sides of gemstone already or not according to the gemstone image information and moves two cutting machines to the two sides of the jade when it is judged that the jade is not exposed out of the two sides of the gemstone; the cutting machines are started until cutting at the time is completed; a first two-claw air cylinder on a first grabbing mechanism grabs the gemstone, a first mechanical arm drives the first two-claw air cylinder to rotate, and the gemstone is placed on a first lifting table after the angle of the gemstone is adjusted; and a second two-claw air cylinder on a second grabbing mechanism grabs the gemstone from a conveying belt, and places the gemstone on a second bearing plate of a second lifting table, the gemstone is fixed to the second bearing plate through two side fixing mechanisms, and the gemstone is polished through a polishing mechanism. The cutting devices and the polishing device are used in cooperation, and therefore maximized removal of gemstone skin and shell is carried out on the premise that the integrity of the interior jade is guaranteed.

The embodiment of the invention discloses a plant irrigation method and device, a storage medium and a faucet. The method comprises the steps of obtaining an irrigation image collected by a camera, inan irrigation range, and recognizing a to-be-irrigated plant in the irrigation image; determining the actual water demand of the to-be-irrigated plant; and irrigating the to-be-irrigated plant according to the actual water demand. According to the method, the actual water demand of the to-be-irrigated plant is intelligently determined on the basis of identification of the to-be-irrigated plant, the faucet can be automatically controlled to irrigate the to-be-irrigated plant according to the actual water demand, the problem that the water quantity is not accurately mastered according to nakedeyes is effectively solved, and manual participation is not needed in the whole irrigation process. The method is more intelligent.

The invention discloses a water cup, a water cup package and a measuring method. The water cup package comprises at least two sensors, a near-field communication module and a data processing module. The sensors are arranged at different positions of the water cup; the near-field communication module is used for receiving measurement data transmitted by the various sensors; the data processing module is used for receiving and processing summarized measurement data transmitted by the near-field communication module. The water cup, the water cup package and the measuring method have the advantages that data can be transmitted and summarized between the sensors at the different positions of the water cup by the aid of near-field communication technologies, and accordingly the measurement dataof the various sensors of the water cup can be processed in a centralized manner.

The invention discloses a display method, a terminal and a storage medium. AR special effects can be well displayed by obtaining input object description information, a to-be-displayed object is determined based on the object description information and the to-be-displayed object is displayed in a preset space, and the problem in the prior art that a user can only display limited AR special effects, the display method is not intelligent enough is solved. According to an AR special effect method provided by the invention, the to-be-displayed object can be determined according to the object description information input in real time, so that various AR special effects can be displayed by a user based on the method provided by the invention, and even the AR special effects designed by the user can be displayed, and therefore, the invention is relatively intelligent. The user can have better interaction experience.

The invention discloses a vehicle uphill and downhill assisting method and system for an electric power assisted shared vehicle. A gyroscope is arranged on the electric power assisted shared vehicle.The vehicle uphill and downhill assisting method for the electric power assisted shared vehicle comprises the steps of S1, collecting angular velocity data of the vehicle by using the gyroscope; S2, judging uphill and downhill states of the vehicle according to the angular velocity data; and S3, if the vehicle is in the uphill state, transmitting an electric power assisting instruction to a battery pack, wherein the battery pack starts electric power to assist the vehicle to go uphills. According to the vehicle uphill and downhill assisting method for the electric power assisted shared vehicle, the angular velocity data of the vehicle is collected by using the gyroscope arranged on the electric power assisted shared vehicle; the uphill and downhill gradients can be accurately judged by using the angular velocity data measured by the gyroscope to judge the uphill and downhill states of the vehicle in the running process; when judging that the vehicle is in the uphill state, the electricpower assisting instruction is transmitted to the battery pack; and the battery pack provides intelligent electric power to the electric power assisted shared vehicle according to the instruction; and therefore, the method is more intelligent and safer, the cost is low, and the riding experience is improved for users of the electric power assisted shared vehicle.

The invention discloses a method for in-situ nondestructive detection of stone sources of stone cultural relics, which is used for detecting the stone sources of the stone cultural relics and comprises the following steps: determining potential sources of stones ofthe detected stone cultural relics through regional geologic features, historical records and the like; utilizing a portable X-ray fluorescence spectrophotometer to test chemical components of the detected stone cultural relics and stones of a potential source of the stone cultural relics; carrying out statistical analysis on the detected chemical component data by respectively utilizing a clustering analysis method and a Bayesian principal component analysis method; when the stones of the detected stone cultural relisc are matched with a certain potential source stone, obtaining the source of the stones of the detected stone cultural relics; and if the stones of the detected stone cultural relics is not matched with the potential source stone, continuously repeating the steps until the detected stone cultural relic stone is matched with the potential source stone. According to the method, the detected stone cultural relics are not damaged, sampling is not needed, the reliability is high, and in-situ detection can be achieved.

The invention provides an enclosed space fire hazard detection alarm system and an alarm method based on an electronic nose technology. The alarm system comprises a sampling pipe, a filter device, three gas sensors, a temperature sensor, a sensor power supply unit, a sensor control unit, a signal collecting unit, a signal conditioning unit, a monitoring host, a sound-light alarm device, a flowmeter, a vacuum pump, a three-way electromagnetic valve and a tail gas processing device. By means of the system and the method, real-time on-line analysis and study and intelligent judgement can be achieved, a fire hazard can be found in a very early stage, false alarm rate and report-failure rate can be reduced, and damage to persons and equipment in an enclosed space of the first hazard can be reduced to the larger extent.