CN209542474U - Conveyer belt multi-material category identification device - Google Patents

Conveyer belt multi-material category identification device Download PDF

Info

Publication number
CN209542474U
CN209542474U CN201821709179.8U CN201821709179U CN209542474U CN 209542474 U CN209542474 U CN 209542474U CN 201821709179 U CN201821709179 U CN 201821709179U CN 209542474 U CN209542474 U CN 209542474U
Authority
CN
China
Prior art keywords
light source
camera
mounting bracket
conveyer belt
fixed
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201821709179.8U
Other languages
Chinese (zh)
Inventor
陈毓昕
顾大同
王仁杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kamimi Yasushichi Automation Technology Ltd
Original Assignee
Kamimi Yasushichi Automation Technology Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kamimi Yasushichi Automation Technology Ltd filed Critical Kamimi Yasushichi Automation Technology Ltd
Priority to CN201821709179.8U priority Critical patent/CN209542474U/en
Application granted granted Critical
Publication of CN209542474U publication Critical patent/CN209542474U/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

A kind of conveyer belt multi-material category identification device, the device includes the mounting bracket for possessing 4 root posts, industrial camera, light source, cooling slag-blowing equipmemt is fixed in mounting bracket, light source forms to obtain frame shape light belt by 4 frames, 4 jiaos of the light source are fixed on 4 columns of mounting bracket, and the light source is located at the lower section of industrial camera, the surrounding of industrial camera is cooling slag-blowing equipmemt spray head, Universal electric control cabinet is fixed near mounting bracket, mounting bracket is located at the top for conveying the conveyer belt of various materials, the camera lens of industrial camera is towards the material on conveyer belt, mounting bracket top and be surrounded by barn door, the mounting bracket is fixed on ceiling girder steel by hanger, for giving camera, the power supply of light source and cooling slag-blowing equipmemt power supply, and it is responsible for control camera, light source, cool down soot blowing The network equipment of controller and the camera image data transmission of equipment is placed in Universal electric control cabinet.

Description

Conveyer belt multi-material category identification device
Technical field
The utility model belongs to steel works equipment technical field, in particular to a kind of conveyer belt multi-material category identification device.
Background technique
Belt Conveying loose unpacked material is the main transport material mode of steel mill, due to can exist coke, iron ore, sinter and The case where various materials such as pelletizing are conveyed because of technique and Equipment through same belt, existing method need manually Mode is judged that the identification and subsequent the measure of production system to material bring difficulty.
Utility model content
The utility model provides a kind of conveyer belt multi-material category identification device, which is to expect for conveyer belt more Kind type online recognition system, conveyer belt multi-material type online recognition system is can not in order to solve existing material conveyer belt The different material variety of automatic identification, and the problem that can only be solved using manual type.
One of the utility model embodiment, a kind of conveyer belt multi-material category identification device.The device includes possessing 4 to stand The mounting bracket of column, industrial camera, light source, cooling slag-blowing equipmemt are fixed in mounting bracket, and light source is made of 4 frames Frame shape light belt is obtained, 4 jiaos of the light source are fixed on 4 columns of mounting bracket, and the light source is located at the lower section of industrial camera, The surrounding of industrial camera is cooling slag-blowing equipmemt, and Universal electric control cabinet is fixed near mounting bracket.
Mounting bracket is located at top for conveying the conveyer belt of various materials, and the camera lens of industrial camera is towards on conveyer belt Material, mounting bracket top and be surrounded by barn door.The mounting bracket is fixed on ceiling girder steel by hanger On.
Power supply for powering to camera, light source and cooling slag-blowing equipmemt, and be responsible for control camera, light source, cooling are blown The network equipment of controller and the camera image data transmission of grey equipment is placed in Universal electric control cabinet.
The conveyer belt multi-material category identification device is used for conveyer belt multi-material type online recognition system.This is System includes host computer, the operation of ipc monitor conveyer belt, and obtains image data, analyzes material variety in real time.System Include: for the material recognition process on conveyer belt
Conveyer belt multi-material category identification device acquires the image of reload on conveyer belt starting, end by industrial camera; Host computer recognizer judges material type, while obtaining system time reloading the time as the material;By reloading for acquirement Time uploads to host computer material metering software module, and is saved in the database of host computer.
The conveyer belt multi-material category identification device of the utility model is in conveyer belt multi-material type online recognition system A critical component so that conveyer belt multi-material type online recognition system can be realized a conveyer belt multi-material and know automatically Not, accurate, efficient, scientific continuous data is provided for subsequent metering link.
Detailed description of the invention
The following detailed description is read with reference to the accompanying drawings, the utility model illustrative embodiments above-mentioned and other Objects, features and advantages will become prone to understand.In the accompanying drawings, it is practical new that this is shown by way of example rather than limitation Several embodiments of type, in which:
Material recognizing device schematic diagram figure in Fig. 1 the utility model embodiment.
1 --- mounting bracket, 2 --- Universal electric control cabinet, 3 --- light source, 4 --- industrial camera, 5 --- cooling is blown Grey equipment, 6 --- ceiling girder steel, 7 --- hanger, 8 --- conveyer belt, 9 --- barn door.
Specific embodiment
According to one or more embodiments, as shown in Figure 1, a kind of conveyer belt multi-material category identification device.The device packet Include the mounting bracket for possessing 4 root posts, industrial camera, light source, cooling slag-blowing equipmemt are fixed in mounting bracket, light source be by 4 frames form to obtain frame shape light belt, and 4 jiaos of the light source are fixed on 4 columns of mounting bracket, and the light source is located at industry The lower section of camera, the surrounding of industrial camera are cooling slag-blowing equipmemts, and Universal electric control cabinet is fixed near mounting bracket.
Mounting bracket is located at top for conveying the conveyer belt of various materials, and the camera lens of industrial camera is towards on conveyer belt Material, mounting bracket top and be surrounded by barn door.The mounting bracket is fixed on ceiling girder steel by hanger On.
Power supply for powering to camera, light source and cooling slag-blowing equipmemt, and be responsible for control camera, light source, cooling are blown The network equipment of controller and the camera image data transmission of grey equipment is placed in Universal electric control cabinet.
Conveyer belt multi-material category identification device is applied to conveyer belt multi-material type online recognition system, which includes Host computer, the operation of ipc monitor conveyer belt, and image data is obtained by identification device, material kind is analyzed in real time Class.
According to one or more embodiments, material recognizing device be may be mounted in open-air atmosphere, and need 24 hours Work.In order to reduce influence of the external environment (dust, illumination, weather) to image analysis, and it can accomplish unattended work Make, designs lightning protection device, power control, dust-extraction unit, radiator, lighting source, system equipment in material identification division Mainly controlled by high-speed color industrial camera, camera lens, light source, industrial personal computer, power supply (air-supply valve, camera, light source power supply), power supply The equipment such as device, bracket and dust cover composition.Wherein,
The image that camera collects material is sent to computer by optical fiber.Light source is to customize positive white light source, for shining The surface of bright material becomes apparent from the surface characteristics of material.Used in sunshade board makes to customize light source in blocking external environment interference light Play maximum effect.There is lightning protection Anti-surging module in Universal electric control cabinet, guarantees the stability of power supply.
There is PLC in Universal electric control cabinet to control the power supply of camera, light source and cooling soot blower system jet, PLC control Equipment worked according to production line state, if production line is ready, open camera light source and cooling soot blower system electromagnetism Valve.
There are five dust prevention sprinklers altogether in the soot blower system that cools down: a camera shield glass dust prevention sprinkler, a camera dissipate Thermal jet head, two light source dust prevention sprinklers, one be viewport dust prevention sprinkler.Computer can send the instruction that stops production to PLC, and halt production refers to PLC after issuing is enabled to close camera fan and air injection equipment.Computer can issue start command to PLC, after start command issues Start all devices.
According to one or more embodiments, during belt transmits material, material variety is identified, the present embodiment is using first Into and mature industrial grade robotic vision technique (machine vision be exactly replace human eye with machine to measure and judge), have developed Loose unpacked material identifying system identifies material kind different on belt according to the characteristics of image for transmitting different materials on belt Class, and the result of identification is transferred to subsequent metering system by cable or optical fiber, metering system easy to produce is to material Accurate measurement valuation.
The maximum width of conveyer belt: 1.47 meters, belt maximum speed: 2m/s, material variety include coke (four meter three it is burnt, Six meters of cokes, high hole is burnt, Jing Dezhen south is burnt, table mountain is burnt, the upper pool is burnt), Jing Gangshan mine, Yunfu mine, South Africa Iron Ores, sintering iron ore, knob The classifications such as graceful powder, pelletizing (good mountain ball, thyrite ball, safe rich ball, Tongling ball, Chilean ball), fluorite.
Multi-material type online recognition method is broadly divided into two steps, i.e. study and identification step.The recognition methods Feature extraction is carried out for material, to guarantee that algorithm can be carried out effective identification.For each material, captured by camera Picture include various features, the difference between these feature instantiations material.It is special that this algorithm is mainly extracted color, texture The features such as intensity of light after matter, geometry, polishing.These feature instantiations essential difference of each material, as coke color is inclined Blue-black, Jinggang Mountain mine color is partially red, sintered surface out-of-flatness, and pelletizing is mainly circle etc..Therefore they can be used to effectively Identify material.
In learning procedure, system acquires the picture of material not of the same race on a moving belt in real time, and is stored in different files Learning sample in folder as this algorithm.This algorithm reads the picture under each leaning portfolio and generates their corresponding marks Label.Then the extraction of feature is carried out such as to every picture: color characteristic, texture speciality, geometry, the intensity of light after polishing Deng.Since these features extracted can not be directly used in identification classification, thus need to handle these features so that The discrimination of different types of material is more obvious.Then calculate the center of a sample of every class sample as this kind of other judgements according to According to.
In identification step, to industrial camera, picture collected carries out feature extraction on a moving belt first, then right These features, which carry out processing identical with study part, can effectively identify institute to guarantee to learn the obtained center of a sample in part The picture of acquisition, the last kind that identified material is judged according to the center of a sample for learning obtained by part every kind of material Class.Since the identification of single does not ensure that discrimination very, guaranteed using multiple step format ballot on conveyer belt The recognition result of a batch of material can reach accuracy rate very, i.e., first voted with multiple recognition result multiple first to generate Grade recognition result, then generates final recognition result to primary recognition result Nearest Neighbor with Weighted Voting.The result quilt finally identified It reloads the time as judgement, generates the foundation of weight of material by the gross.
Identifying system acquires belt material using high speed high clear colorful area array cameras in real time, according to material variety on belt Difference, it is for statistical analysis to characteristics of image such as shape, the colors shown, it can realize that non-contact material is known using machine vision Not.
In the center vertical direction position mounting industrial camera of belt, real time image collection point is carried out to belt movement process Analysis.A kind of status information of the variation (no material is also considered as material) of camera analysis belt filled is sent in the operational process of belt To host computer, record camera uploaded state information and image information are analyzed by host computer, and in material metering software automatically with Plan matching.
It when material being identified and measured using identification device, lays a course first, the material kind for listing plan feed is suitable Sequence, feed information need to include material major class (such as pelletizing, iron ore, coke etc.) and its detailed name of an article (such as thyrite ball, good ball etc.).Object Material identifies each period material transportation situation by identification device when transporting on weighing belt, by material recognition system, in order Auto-matching is effectively fed plan.If the major class identified is consistent with plan major class, indicate that the plan is completed, and the plan of taking out In the detailed name of an article generate complete material transferring information (including period, the information such as material kind), issue on webpage.Both sides are single Position confirmation, it is errorless, confirmation message is submitted, the modification information if wrong.After material information is confirmed, by weighman's personnel's core To metrical information, generates complete report and externally issue and checked for other unit.
It is worth noting that being created although foregoing teachings describe the utility model by reference to several specific embodiments Spirit and principle, it should be appreciated that, it is not limited to the specific embodiments disclosed for the utility model, draws to various aspects Divide and does not also mean that the feature in these aspects cannot combine, it is this to divide the convenience merely to statement.The utility model purport Covering various modifications and equivalent arrangements included in spirit and scope of the appended claims.

Claims (1)

1. a kind of conveyer belt multi-material category identification device, which is characterized in that the device includes the installation branch for possessing 4 root posts Frame, industrial camera, light source, cooling slag-blowing equipmemt are fixed in mounting bracket, and light source forms to obtain frame shape light by 4 frames Band, 4 jiaos of the light source are fixed on 4 columns of mounting bracket, and the light source is located at the lower section of industrial camera, industrial camera Surrounding be cooling slag-blowing equipmemt, Universal electric control cabinet is fixed near mounting bracket,
Mounting bracket is located at top for conveying the conveyer belt of various materials, and the camera lens of industrial camera is towards the object on conveyer belt Material,
Mounting bracket top and be surrounded by barn door,
The mounting bracket is fixed on ceiling girder steel by hanger,
Power supply for powering to camera, light source and cooling slag-blowing equipmemt, and be responsible for control camera, light source, cooling soot blowing and set Standby controller and the network equipment of camera image data transmission are placed in Universal electric control cabinet,
The slag-blowing equipmemt that cools down includes 5 dust prevention sprinklers: 1 camera shield glass dust prevention sprinkler, 1 camera heat dissipation spray-head, 2 Light source dust prevention sprinkler, 1 be viewport dust prevention sprinkler.
CN201821709179.8U 2018-10-22 2018-10-22 Conveyer belt multi-material category identification device Active CN209542474U (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201821709179.8U CN209542474U (en) 2018-10-22 2018-10-22 Conveyer belt multi-material category identification device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201821709179.8U CN209542474U (en) 2018-10-22 2018-10-22 Conveyer belt multi-material category identification device

Publications (1)

Publication Number Publication Date
CN209542474U true CN209542474U (en) 2019-10-25

Family

ID=68247092

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201821709179.8U Active CN209542474U (en) 2018-10-22 2018-10-22 Conveyer belt multi-material category identification device

Country Status (1)

Country Link
CN (1) CN209542474U (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112356303A (en) * 2020-10-22 2021-02-12 深圳市能迩环保科技实业发展有限公司 Brick making raw material recognizing and proportioning device

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112356303A (en) * 2020-10-22 2021-02-12 深圳市能迩环保科技实业发展有限公司 Brick making raw material recognizing and proportioning device

Similar Documents

Publication Publication Date Title
CN106362957B (en) A kind of gangue piece-rate system and separation method
CN104101608B (en) Intelligent detecting device for detecting defects of multi-type irregularly shaped product
CN109060827A (en) A kind of intelligent visual detection identification equipment
CN105181709A (en) Combined multiple-light-source on-line detection device and method for appearance defects of brake pad
CN102063746B (en) Program-controlled patrol machine and program-controlled patrol method for transformer station
CN105651782A (en) Engine cylinder cover defect automatic detection equipment based on machine vision
CN105717132B (en) A kind of the built-in fitting detection system and method for PC component production line
CN205484100U (en) Flaw off -line monitoring system of cloth
CN109000564A (en) A kind of keycap assembling method for testing performance and automatic detection device
CN209542474U (en) Conveyer belt multi-material category identification device
CN104458750A (en) Automatic aluminum profile surface defect detecting equipment based on machine vision
CN108627520A (en) A kind of on-line detecting system and method for heterogeneous solid material presentation quality
CN106000912A (en) Citrus cyst foreign matter removing system based on machine vision guidance
CN111805508B (en) On-site security permission method based on visual understanding
CN108809258A (en) A kind of photovoltaic module hot spot detecting system and hot spot detection method
CN108566538A (en) Based on the circular coal yard personnel safety guard of Infrared-Visible fusion tracking and the monitoring system and method for spontaneous combustion
CN205879493U (en) On -line measuring system of car light lens
CN109772721A (en) Engine commutator open defect automatic online vision inspection apparatus and detection method
CN110057198A (en) Wheel of sintering trolley working state detecting method and detection device
CN208366871U (en) Detection system
CN112365015A (en) Robot collaborative operation and maintenance monitoring method based on RFID active positioning
CN109759755A (en) AI intelligent process anomalous identification processing system and solar cell chip bonding machine
CN206370163U (en) A kind of aircraft berths visual Intelligent boot device
CN106000903B (en) A kind of non-contact detection system and method for being used to detect loom cloth gripper edge gap
CN205942172U (en) Positioner and display screen that display screen detected machine detect machine

Legal Events

Date Code Title Description
GR01 Patent grant
GR01 Patent grant