50results about How to "Avoid missing defects" patented technology

This invention discloses one method and device to test fruit deficiency based on multi-sensor information integration, which comprises the following steps: using multiple testers to get light source to light fruit different parts spectrum information; test mode adopts continuous remove method and K near method or their combination and referring to fruit environment temperature and humidity factors to identify deficiency. This invention method uses multiple testers to test fruit to avoid leakage.

The invention discloses a liquid crystal screen Mura defect detection method based on computer vision. The method comprises the following steps: acquiring an image at an inclined visual angle; extracting a region of interest based on a fixed threshold method and carrying out perspective transformation based on a perspective transformation method, image texture filtering based on a Gabor filter, image color space conversion, Mura defect segmentation based on a sliding window scanning mode of a dynamic change size and a self-adaptive double-threshold segmentation algorithm double mechanism, anddefect segmentation area secondary screening. According to the liquid crystal screen Mura defect detection method based on computer vision, the problems of missing detection and lack of consistency ofdetection results due to visual fatigue easily caused by long-time manual detection of Mura defects can be solved, the detection efficiency of the Mura defects is improved, and the consistency of thedetection results is ensured.

The invention relates to the technology of plastic pipe electric melting joint detection and aims at providing an ultrasonic detection method for a defect automatic identification auxiliary electric melting joint smart phased array. The ultrasonic detection method comprises the steps of carrying out joint defect detection preparation, carrying out joint defect detection imaging and carrying out joint defect detection identification. Welding defect identification is realized through utilization of a defect automatic identification module based on a neural network algorithm. The welding defect identification comprises the content of preprocessing welding defect detection image data, carrying out classification identification on feature areas based on the neural network algorithm and carryingout safety assessment on typical defects. According to the method, typical defect automatic identification, real-time display and alarm prompt in a detection process are realized, dependence of a traditional phased array ultrasonic detection method on a qualification of a detector is overcome, and defect missing detection can be effectively avoided. According to a defect automatic identificationtechnology based on the neural network algorithm, operation speed is fast, complex defect identification accuracy is high, computing load of an ultrasonic detecting instrument can be reduced, and defect detection and identification are guaranteed to be displayed synchronously in real time.

The invention relates to the fields of hot-dip galvanizing alloy technology and auxiliary equipment used therefor, and specifically relates to a dry and wet gas mixing and humidifying device for inhibiting zinc ash in hot galvanizing and a use method thereof. The device comprises a furnace nose (10), and is characterized by also comprising a water tank (1), a water inlet pipe (21), a water discharge pipe (22), an air inlet pipe (23), a flow meter (33), a water level gauge (41), a heater (42), a dew-point instrument (43), a dry gas header pipe (5), a dry wet gas mixing pipe (6), and a control device (7). The use method comprises the following steps in sequence: (i) introducing dry gas; (ii) humidifying the dry gas; (iii) mixing dry gas and wet gas; and (iv) regulating the dew point. The device and method have the advantages that the zinc ash is well inhibited, the strip steel quality is high, the control is accurate, and the automation degree is high.

The embodiment of the invention discloses a data processing terminal for a tire. The terminal comprises a first data receiving and transmitting unit and a data input unit; a worker can input a tire transfer record of some tire through the data input unit at any time, wherein the first data receiving and transmitting unit is capable of achieving data exchange with a data monitoring module arranged in the tire, a cloud server and a central control machine put in a vehicle, so that the worker can obtain identification information of some tire through the terminal, inputs the tire transfer record or other related data of the tire, and transmits the identification information and the input data to the cloud server, so that the recording process is simple and the problem of data loss easily caused in the manual recording process of the worker is avoided.

The invention relates to a front-end probe mechanism of a reactor pressure vessel checking machine. The front-end probe mechanism comprises an ultrasonic probe body, a probe support framework framed outside the ultrasonic probe body, and a wall support frame in rotary connection with the probe support framework through a third rotating shaft; the probe support framework comprises an inner frame in rotary connection with two opposite sides of the ultrasonic probe body through a first rotating shaft respectively and an outer frame in rotary connection with two opposite sides of the inner frame through a second rotating shaft respectively; a torsion spring is also sleeved on the third rotating shaft; one end of the torsion spring is connected with the wall support frame, while the other end is connected with the outer frame; and the first rotating shaft, the second rotating shaft and the torsion spring provides a rotating freedom for the ultrasonic probe body on a spherical surface so that the ultrasonic probe body is tightly coupled with the spherical surface to be detected. The probe mechanism overcomes the defect of detection data loss due to poor coupling of the probe body and the spherical surface, and lays a foundation for safe operation of a nuclear power station.

An ultrasonic time-of-flight diffraction method by a cylinder focusing wedge relates to the field of ultrasonic detection and solves the problem that an existing time-of-flight diffraction method for detection is low in energy level of defective diffraction signals and needs improvement of detecting sensitivity. The ultrasonic time-of-flight diffraction method includes the specific steps: firstly, determining known conditions including the diameter of a probe wafer, the longitudinal wave velocity of the wedge, the longitudinal wave velocity of a detected workpiece and the velocity of longitudinal waves in water; secondly, setting needed conditions including the water path of an acoustic beam of a spindle, the refraction angle of longitudinal waves in the detected workpiece and the focusing vertical depth of the acoustic beam of the spindle in the workpiece; and thirdly, computing the curvature radius of a cylinder of the wedge according to the conditions and detecting by the ultrasonic time-of-flight diffraction method through the manufactured cylinder focusing wedge. The ultrasonic time-of-flight diffraction method is used for ultrasonic detection.

The invention relates to a data acquisition method and electronic equipment. The method comprises the following steps: receiving a map data acquisition task sent by a server, wherein the acquisition task comprises target area information; triggering the map data acquisition task after entering the target area is judged according to the target area information; acquiring map data in the target areaaccording to the map data acquisition task; and after it is judged that the user leaves the target area according to the target area information, sending the collected map data in the target area tothe server. According to the scheme provided by the invention, the integrity of map data reporting can be ensured.

This invention is about a dienergetic hard electron (whose energy range is 10-1400keV) three-dimension volume spectrum image-forming device, which contains removable dienergetic X-ray source orª†-radioisotope sources inside the system and the suited X-ray plate detector. The light source and the detector have a mobile linkage on a rotatable bracket, and between them there is a space for monitoring substance. The source move track can be a closed curve composed of two parallel arcs and two parallel beelines separately linking both ends of the two arcs. Its radial beam is a prick beam that always covers the whole monitoring substance, and faces to detecting side of the detector and forms monitoring substance projection on the detecting side. By using this invention, the high-resolution and isotropical three-dimension data set of soft and hard tissue on inspection personnelí»s unusual position can be obtained, and do image rebuilding image processing through computer system to form the three-dimension volume image. It can be used to physic examine flesh or inspect inner structure of industrial goods.

The present invention provides a single cell whole genome sequencing method and an application of an amplification method in constructing a gene library and conducting single molecule sequencing. Theamplification method comprises step of contacting nucleic acid with nucleic acid polymerase and random primers to carry out multiple displacement amplification of the nucleic acid and strands.

The invention discloses a UWB label upgrading method and system based on ultra wide band wireless communication. The UWB label comprises an area A used for storing BootLoader, an area B used for storing an upgrading mark, an area C used for configuring parameters and an area D used for storing a terminal program. UWB label upgrading comprises the following steps: receiving an upgrading instructionissued by an upper computer, and switching the UWB label from the area D to the area A; obtaining an upgrade package issued by the upper computer, and writing upgrade package data into the area D; when receiving of the upgrade package is completed, writing an upgrade completion mark into the area B; and restarting the UWB label, and completing upgrading of the UWB label. According to the UWB label upgrading method, online software upgrading of the UWB label is achieved on the premise that the cost is guaranteed, and the maintainability of the system is improved.

The invention discloses a comprehensive malicious code detection system and method in an electric power production system, and the method comprises the steps: setting anti-malicious code client software on a plant station terminal in a plant station, and enabling the anti-malicious code client software to be used for searching and killing malicious codes on the plant station terminal and forming a historical record message; a management large area acquisition machine is arranged in the production control large area; the plant station terminal is connected with the management large area acquisition machine; the plant station terminal and the management large area acquisition machine are connected with the attached server; the defects that in the prior art, historical record messages formed by malicious code prevention client software are only stored locally and are not transmitted to a management region acquisition machine, so that the messages acquired by the management region acquisition machine are not complete, and the historical record messages are omitted during transmission due to message congestion of the management region acquisition machine are effectively avoided.

The invention provides an image processing method and device. The method comprises the steps: firstly obtaining a first low-illumination image; then, images of the first low-illumination image under different exposure proportions are generated, and a plurality of first derivative images are obtained; then, respectively extracting effective image feature information in the first low-illumination image and each first derivative image; and finally, cascading the effective image feature information in the first low-illumination image with the effective image feature information in each first derivative image, and generating a first target image corresponding to the first low-illumination image based on the cascaded effective image feature information. According to the technical scheme, the clear target image is determined through the low-illumination image and the derivative images of the low-illumination image under different exposure proportions, so that the defect of low sensitivity tothe exposure proportion of the low-illumination image in the prior art is overcome, and the defect of loss of image details can be effectively avoided.

The invention provides a semiconductor wafer test method, and a wafer projection manufacturing method, a semiconductor device and an electronic device and relates to the technical field of a semiconductor. The semiconductor wafer test method comprises a step that during a probe test, test probe mark is made to deviate from the center position of a welding pad. The wafer projection manufacturing method comprises steps that during the probe test, the test probe mark is made to deviate from the center position of the welding pad, is utilized at a welding pad opening forming a wafer projection and is arranged at a region not influenced by the test probe mark. The semiconductor wafer test method and the wafer projection manufacturing method are advantaged in that influence of the test probe mark on manufacturing a ball bottom metal layer in the projection process can be completely prevented, a defect of projection missing can be avoided, packaging reliability is improved, and the packaging yield is further improved. The semiconductor device and the electronic device employing the semiconductor wafer test method and the wafer projection manufacturing method have similar advantages.

The invention provides a bottle bottom defect detection system and method. The system comprises a light source, a reflector array, a light collecting device, a shooting device and a control device; the light source irradiates the bottom of a to-be-detected product at a detection position; the reflector array receives a plurality of beams of return light carrying bottom information and reflected bya plurality of equally-divided areas at the bottom, and transmits the return light reflected by each equally-divided area to the light collecting device; the light collecting device reflects the plurality of beams of return light into the visual field range of the shooting device in a non-overlapping manner; the shooting device generates a bottom image comprising each equally-divided area according to the plurality of beams of collected return light, and sends the bottom image to the control device; and the control device identifies the bottom defects of the product according to the bottom image and outputs a result. The problems of low human eye visual detection speed, low detection precision and error proneness are overcome.

The invention relates to the technical field of image processing, and discloses an image processing method and system in a highway technical condition evaluation system so as to improve efficiency and reliability. The method comprises the following steps: acquiring continuous image sets of measured highway mileages according to lanes in a time-sharing manner; determining a corresponding reference background grid for each target image; performing subtraction on each grid image in the target image and an image corresponding to the reference background grid, then calculating whether the variance of the gray sequence of each pixel point in the grid image after subtraction is greater than a set second threshold value, and if yes, determining the corresponding grid as a potential disease grid; fusing adjacent potential disease grids into the same potential disease area, automatically marking boundary information of each potential disease area, and calculating a corresponding area; and according to the area of each potential disease area and the pavement material of the target image, pre-screening a disease list corresponding to each potential disease area in a one-to-one manner for a user to manually input a disease type and then store the disease type.