CN105195438A - Embedded type automatic pearl sorting device and method based on image recognition - Google Patents
Embedded type automatic pearl sorting device and method based on image recognition Download PDFInfo
- Publication number
- CN105195438A CN105195438A CN201510617401.6A CN201510617401A CN105195438A CN 105195438 A CN105195438 A CN 105195438A CN 201510617401 A CN201510617401 A CN 201510617401A CN 105195438 A CN105195438 A CN 105195438A
- Authority
- CN
- China
- Prior art keywords
- pearl
- sorting
- image
- image recognition
- embedded automatic
- 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.)
- Granted
Links
Landscapes
- Image Processing (AREA)
Abstract
The invention provides an embedded type automatic pearl sorting device and method based on image recognition. A pearl image is collected through a camera and processed through a kernel processing module, and the size of pearls is obtained through calculation of the proportions of actual geometric area corresponding to pixel points; the colors of the pearls are obtained through the RGB values of effective pixels of the pearls in the image; circle center calculation is replaced with pearl boundary gravity center calculation, the maximum radiuses and the minimum radiuses of target are found through a traversing method so as to judge the roundness of the pearls, then a sorting control command is output so that a pearl sorting mechanism can automatically sort the pearls according to set sorting standards, and sorting result data are saved. The embedded type automatic pearl sorting device and method based on image recognition can sort jumble pearl samples according to the set standards, the sorted pearls are automatically conveyed into different grooves, and meanwhile sorting results are transmitted to upper computer monitoring software to be displayed.
Description
Technical field
The present invention relates to a kind of sorting equipment, especially relate to a kind of embedded automatic pearl sorter based on image recognition and method for sorting.
Background technology
In current pearl processing industry, pearl sorting, be nearly all by the backward situation manually completed, and this field does not have the present situation of ready-made automatic sorting equipment for pearls both at home and abroad.Pearl is as nonplanar spheroid target, surface has certain radian, will according to size, shape, bright and clean, flaw and color grading, the image of the whole spherome surface of pearl must be obtained, therefore, detect a pearl and to its carry out effective classification need obtain pearl image and to its analyze after just can carry out classification.The process of native system combining image and automated control technology have made a kind of novel, small and exquisite, inexpensive, automatic pearl sorter.
Summary of the invention
The invention provides a kind of embedded automatic pearl sorter based on image recognition and method for sorting, solve the problem of pearl sorting, its technical scheme is as described below:
A kind of embedded automatic pearl sorter based on image recognition, comprise pedestal, and core processing module and coupled image capture module, sorting executing agency, alternating interface between man and computer, data store and monitor control mechanism, described image capture module adopts colour imagery shot, and described sorting executing agency comprises steering wheel for loading and discharge pearl particles, for pearl being moved to correct classification position reducing motor.
Also be provided with light compensation device, described light compensation device comprises light harvester and light output-controlling device, described light harvester comprises equally distributed photo resistance, described light output-controlling device comprises the LED evenly arranged, for compensating field rays, regulate light intensity by regulating the PWM output pulse width of core processing module.
Described alternating interface between man and computer comprises the single-chip microcomputer and color screen LCD that are connected by FSMC interface, and button and touch-screen.
Being provided with positioning strip above described sorting executing agency, by arranging the state of infrared tube detection and location bar, being obtained the position of pearl mechanism for sorting by the state reading infrared tube.
Described sorting executing agency also comprises the litter being provided with tooth bar, for coordinating the rotation of reducing motor.
Based on an embedded automatic pearl method for sorting for image recognition, comprise the following steps:
(1) gather pearl image by image capture module, described image capture module adopts colour imagery shot;
(2) through core processing module, image is processed, judge the size of pearl, color, circularity, output sort control command;
(3) make pearl mechanism for sorting automatically classify to pearl according to the criteria for classification of setting, and the result data of classification is preserved.
Further, in step (2), image processing method is as described below,
1. the size of pearl is calculated by the ratio of the corresponding actual geometric area of the pixel of camera;
2. from image the valid pixel of pearl rgb value in obtain the color of pearl;
3. calculating pearl border center of gravity replaces the center of circle to calculate, and finds the minimum and maximum radius of target, judge the circularity of pearl with this by traversal.
Further, step 1. in, described colour imagery shot carries out IMAQ to fixed size region, makes each pixel of camera and region area have fixing proportionate relationship.
Further, step 1., 3. in calculate the size and shape of pearl, the coloured image of described pearl needs to carry out RGB successively and turns gray value, image filtering, image binaryzation process, edge detection, field, edge eight tracking step.
Further, above pearl mechanism for sorting, be provided with positioning strip, and be provided with the infrared tube of detection and location bar state, by the position of infrared tube location pearl mechanism for sorting, the reducing motor controlling pearl mechanism for sorting moves left and right.
The described embedded automatic pearl sorter based on image recognition and method for sorting, the standard of the pearl sample mixed according to setting can be realized to classify, and the pearl of sorting is distributed in different grooves automatically, sorting result is delivered to host computer monitoring software display simultaneously.
Accompanying drawing explanation
Fig. 1 is the functional block diagram of the described embedded automatic pearl sorter based on image recognition;
Fig. 2 is the structural representation of the described embedded automatic pearl sorter based on image recognition;
Fig. 3 is the structural representation of described sorting executing agency;
Fig. 4 is the flow chart of the described embedded automatic pearl method for sorting based on image recognition;
Fig. 5 is the schematic diagram of described eight connectivity Contour extraction method.
Detailed description of the invention
The described embedded automatic pearl sorter based on image recognition, comprise core processing module and coupled image capture module, sorting executing agency, alternating interface between man and computer, data storage and monitor control mechanism, Fig. 1 is the described functional block diagram based on the embedded automatic pearl sorter of image recognition.
Wherein, image capture module uses OV7670 colour imagery shot and STM32F407 single-chip microcomputer composition.Because native system needs to identify the color of pearl, require that camera can identification colors.The interface of this camera and single-chip microcomputer is SCCB interface and DCMI8 bit pattern interface.SCCB is the interface of Single-chip Controlling camera, and single-chip microcomputer, by the register of this interface configuration camera, makes the camera pattern that arranges as required export.DCMI is the utilizing camera interface that STM32F407 provides, and the design adopts 8 bit patterns.
The DCMI interface of described single-chip microcomputer is connected with camera, is controlled camera by SCCB agreement, makes its output system reach the picture format of needs.In order to make, the picture pixels that collects is more, quality is higher.Camera is configured to the colored output format of 565RGB.The pixel of returning according to cam feedback in DCMI interface is interrupted, row interrupts and field interrupts being separated picture frame.Use the DMA passage of single-chip microcomputer that image data is sent in storing mechanism.When image stores, the design is the form of JPEG picture-storage.Consider that when system the quality of the distribution of light to IMAQ has a significant impact, therefore native system have employed the real-time light feedback compensation mechanism of image when designing, and the change reducing extraneous light impacts IMAQ.
Core processing module refers to the modular circuit be made up of STM32F407 single-chip microcomputer and its peripheral circuit, mainly plays the control of each peripheral hardware of process and the effect of image procossing output in systems in which.STM32F407 is based on Contex-M4 framework band FPU single-chip microcomputer, has aboundresources, the advantages such as fast operation, therefore selectedly does control core chip, system cloud gray model dominant frequency 168MHz.
Alternating interface between man and computer is made up of the single-chip microcomputer connected by FSMC interface and color screen LCD and three touch key and touch-screen.Its Main Function is in systems in which display system state and adjustment parameter.When system cloud gray model, the image gathered is shown on liquid crystal display in real time, Real Time Observation can gathers the quality of image.Liquid crystal display LCD is connected with single-chip microcomputer by FSMC interface, and the full name of FSMC is " static storage controller ".This interface is set in STM32F407 single-chip microcomputer and facilitates user's outer extension memory etc., in TFT liquid crystal, utilize the buffering area that RAM shows as data.The object upgrading liquid crystal display is reached by the data reading and writing this buffering area.
Data storage mechanism forms by with the SD card that single-chip microcomputer is connected by SDIO interface.Mainly play the effect that image file stores and sorting outcome history data store in systems in which.Data in the buffering area of DCMI interface directly can be sent to the buffering area of SDIO by the DMI passage of single-chip microcomputer, and then utilize SDIO by deposit data in SD card.In order to file control data more easily.This has been transplanted system keil and has carried Fatfs file management system in system RTX.
Monitor control mechanism is in the sorting result of pearl sorter write SD card, passes through RS232 interface again sending computer in data, utilize host computer monitoring software to generate sorting result form when system connects monitoring host computer.This is conducive to the various rank of pearl in enormous quantities in productive life, containing quantitative statistics, provides Data support for improving pearl quality below.
Single-chip microcomputer gathers image information by DCMI interface, obtains pearl rating information after analyzing and processing.Rating information according to pearl exports control command to stepper motor, steering wheel and reducing motor, realizes pearl automatic sorting.The position of sort positions feedback for locating current executing agency, the corresponding unique outgoing position of pearl of each grade, realizes the separation of different brackets pearl with this.
Be provided with light compensation device in system, it comprises light harvester and light output-controlling device two parts circuit.In the design, light harvester uses four equally distributed photo resistance.By receiving Chip Microcomputer A/D input after a biasing resistor, by reading the power judging light of Chip Microcomputer A/D value.Light output-controlling device is made up of the LED of four high lights, is received the PWM output of single-chip microcomputer by a Darlington transistor ULN2003, regulates light intensity by regulating the PWM output pulse width of single-chip microcomputer.
Automatically adjust the light of image f iotaeld-of-view by adding light closed-loop control, make system gather image at relatively uniform, stable light, high-quality image is the strong guarantee accurately identifying pearl size, shape and color.LCD display and 3 touch keys are the human-computer interaction mechanism of system.
As shown in Figure 2, the described embedded automatic pearl sorter based on image recognition comprises liquid crystal display 1, the cylinder 2 of pearl conveyer belt, stepper motor 3, core processing module 4, camera 5, pearl conveyer belt 6, conveyer belt carrying platform 7.During system works, stepper motor 3 rotation makes the pearl of conveyer belt 6 move forward and enters IMAQ visual field, and the image that core processing module 4 is gathered by process can know that pearl enters the situation of visual field.If there is pearl to enter visual field, stepper motor 3 stops operating, and obtains pearl class information after network analysis.Process rear stepper motor 3 to start, pearl is delivered in pearl mechanism for sorting.
As shown in Figure 3, in the schematic diagram of described sorting executing agency, wherein 8 is reducing motor positioning strip, its underpart is provided with ossificational antler, and 11 is reducing motor, and bottom is provided with steering wheel, 12 is pearl classification lattice, 9 is the litter of diameter 3mm, and 10 is the litter of the tooth bar of 0.5 for modulus, and above each device is positioned on pedestal.During system works, first pearl sorting executing agency is in default conditions, and namely reducing motor 11 is in mid point, and ossificational antler upward.Then reducing motor 11 rotates, and tooth bar is formed horizontal line horizontal movement, makes pearl mechanism for sorting reach the classification position of pearl, and servo driving ossificational antler rotates 180 degree and pearl is fallen, and a Machine cycle completes.
This pearl sorter, by the position of three infrared tube location pearl mechanism for sorting, is divided into 7 classes to need the position of definition more than 7 pearl.Above mechanism for sorting, positioning strip is provided with during native system design.Eight states of detection and location bar need 3 infrared tubes, are obtained the position of pearl mechanism for sorting by the state reading infrared tube.Thus control moving left or moving right of reducing motor.
Fig. 4 is method for sorting of the present invention, first system initialization, determines whether pearl and enters visual field, light compensation is carried out to the pearl entering visual field, then after acquisition pearl image, carry out image procossing, calculate size, the CF of pearl, carry out sorting process by sorting actuating unit.
The present invention utilizes colour imagery shot OV7670 to carry out IMAQ to fixed size region, makes each pixel of camera and region area have fixing proportionate relationship.Calculate the quantity that pearl accounts for image slices vegetarian refreshments, passing ratio relation correspondence obtains shape and the size of pearl.By the RGB ratio value of reading images, obtain the color of pearl.Achieve and also have color to identify to pearl size, shape, according to the standard of setting, pearl is divided into several grades.
Described light compensation controls to be because light has a great impact the quality gathering image, therefore this pearl sorter employs initiatively light compensation when designing, object regulates camera visual field light intensity that each two field picture gathered is under similar environment, thus ensure the quality of pearl image.
Use equally distributed four photo resistance to collect field rays situation in this sorter, utilize the high light LED light filling of four tunable optical.Make service system light control to form closed-loop control, reach and control light effect preferably.In software control, by experiment, can obtain intensity control when 1200Lux, the picture quality gathered is best.
Utilize single-chip microcomputer to read the AD value of photo resistance, darker by the larger light of circuit design known AD value, AD value corresponding when light intensity is 1200Lux is 800.What the design adopted is that pid algorithm controls light intensity, the AD value that AD value sets desired value and collection is compared and obtains optical aberrations, error is substituted into I below
Ratio.
Wherein Kp is scale factor, and Ki is integrating factor, and Kd is differential divisor, and e (n) is error current, and e (n-1) is history error.
At present, to classify the general size according to pearl particles, circularity and color to pearl in market.To be GB/T18781-2008 standard describe about pearl ranking score sector of breakdown following table, and when can obtain the Shape Classification standard of pearl from table according to diameter difference percentage, it is defined as: the average ratio of maximum gauge and minimum diameter.
It is as shown in the table for pearl shaped parameter:
Therefore in order to calculate the shape of pearl, needing the marginal information obtaining pearl, calculating the center of circle, utilize the center of circle to obtain maximum gauge and minimum diameter to each point distance at edge.Both can obtain the size of pearl after carrying out Conversion of measurement unit by the pixel number calculating the pearl image after binaryzation, obtain the rgb value of multiple pixel in original image pearl average after obtain the colouring information of pearl.
System camera collection to image be colored, conveniently calculate pearl shaped size.RGB color image is needed to be converted into gray value.The valid pixel of described camera is 640*320, and in this sorter, the ratio of image pixel number and pearl size is 196: 1.
The object of image filtering is the noise of filtering image, usually uses airspace filter or frequency domain filtering.Airspace filter refers to and carries out filtering to original image according to certain compute mode, and frequency domain filtering then needs carrying out filtering by image conversion to frequency domain.The noise that image in this pearl sorter occurs usually has: data transmit the thermal noise of undesirable existence, pcb bus interference exists white noise, light intensity change the salt-pepper noise caused.Consider, the present invention adopts filter window to be that the median filtering algorithm of 3*3 carries out filtering to pearl original image.Its computational methods are as follows:
(1) gray value of the pixel of filter window is arranged in order;
(2) using the value of the median after arrangement as filter window intermediary image vegetarian refreshments;
(3) mobile filter window is until entire image completes.
Image binaryzation is mainly in order to separate target image from image.Normally choose specific threshold value t, computing once carried out to image:
F (i, j) is (i, j) the individual pixel after binaryzation, and F (i, j) is the gray value of corresponding original image.Can see that threshold value is the key factor of effect diagram as separating effect, after which determining process, whether objective contour truly reflects the objective contour of original image, and as threshold value is excessive, the profile after computing can reduce, and threshold value is too small, and the objective contour after computing expands.In image procossing, the method for selected threshold generally has: state method (peak valley method), judgement, best entropy automatic threshold method the minimum error method etc.For native system, pearl is placed in the background at the blue end, and the color of pearl is generally white, black or other is variegated, assembles so target image and background image can form pixel.So select peak valley method as calculated threshold herein.
Edge is the set that in image, pixel has the pixel of Spline smoothing or the change of roof shape.So can be found by gradient operator.Conventional gradient operator has Roberts gradient operator, Prewitt operator, Sobel operator, Laplace operator, Ma Er operator, Canny edge detection operator etc.The present invention adopts First-order Gradient operator, and this algorithm is simple and reliable.
Edge tracking carries out track record to the coordinate of the edge pixel detected, and obtains the side information of target.Edge detection has reptile method, raster scanning method, eight connectivity Contour extraction method usually.Adopt herein and obtain based on image direction eight connectivity Contour extraction method the side information that eight neighborhood directional ring gets pearl, thus calculate the center of circle of pearl.For the target image of continuum boundary, utilization can describe direction of curve, provides the initial search point of eight neighborhood.The center being eight neighborhood with current pixel of the principle of eight neighborhood directional ring, each pixel of eight neighborhood correspond to a chain code.As shown in Figure 5, A is current pixel, and A0, A1, A2, A3, A4, A5, A6, A7 are the pixel that this pixel is adjacent.In figure, arrow just represents eight chain codes of this pixel.Being defined from continuous print, can not there is sudden change in the chain code of each Searching point, therefore, can obtain following search plan:
(1) from left to right, from top to bottom first of searching image be not the point of 0.And the coordinate information recording this point is Sk (i, j) k=0; Start x=6;
(2) take Sk as the central point in eight fields, take Ax as initiating searches point, counterclockwise search is not the point of 0.Write down the relative position n with Sk;
(3) judge that whether the coordinate of Sk is identical with S0.If so, this search terminates;
(4) positional information of Sk+n is recorded, k=k+n.And the starting point of search next time is determined according to the value of n, corresponding relation is, x=n+6, if x deducts 8 more than the value of 7, x, returns step (2).
Then realize the sorting n-back test of pearl, from system, first obtain the sorting information of pearl, read the position that pearl mechanism for sorting is current, pearl mechanism for sorting is moved to default location.System drive pearl IMAQ visual field conveyer belt is sent to pearl in the box before pearl sorting executing agency steering wheel.Drive reducing motor to move to left according to sorting information or move to right and reach the sort positions of pearl needs, drive steering wheel rotated down pearl, a sorting circulation completes.
Visible, single-chip microcomputer is stored into the view data gathered SD card and is shown in LCD, carries out intercepting the size, the shape that calculate pearl after effective image, denoising, binaryzation etc. operate, obtain the color of pearl to image, and according to the standard set by pearl classification.And operate the reducing motor of pearl mechanism for sorting and the pearl mixed is classified according to grade by steering wheel automatically.And by RS232 serial ports by sending computer host computer on classification results.In image acquisition process, in order to avoid extraneous light has an impact to picture quality, the design utilizes photo resistance to feed back the light around pickup area.Use high-brightness LED lamp to compensate field rays, make field rays a metastable value, ensure that the image gathered is effective, reliable.
The present invention utilizes single-chip microcomputer collection, process pearl figure obtains pearl grade, according to the criteria for classification set by pearl automatic classification.Demonstrate pearl size, feasibility that shape recognition algorithm is run in Single Chip MC in Embedded System, make machine replace manual sorting pearl to become possibility.Novelty of the present invention is: devise a set of ratio more complete pearl sorting frame for movement; Select single-chip microcomputer as the core cell of image procossing; The study of pearl recognizer, research and transplanting; Increase image f iotaeld-of-view light compensation function; With the addition of ipc monitor, sorting result statistical function, make sorter work hommization more.
Claims (10)
1. the embedded automatic pearl sorter based on image recognition, it is characterized in that: comprise pedestal, and core processing module and coupled image capture module, sorting executing agency, alternating interface between man and computer, data store and monitor control mechanism, described image capture module adopts colour imagery shot, and described sorting executing agency comprises steering wheel for loading and discharge pearl particles, for pearl being moved to correct classification position reducing motor.
2. the embedded automatic pearl sorter based on image recognition according to claim 1, it is characterized in that: be also provided with light compensation device, described light compensation device comprises light harvester and light output-controlling device, described light harvester comprises equally distributed photo resistance, described light output-controlling device comprises the LED evenly arranged, for compensating field rays, regulate light intensity by regulating the PWM output pulse width of core processing module.
3. the embedded automatic pearl sorter based on image recognition according to claim 1, is characterized in that: described alternating interface between man and computer comprises the single-chip microcomputer and color screen LCD that are connected by FSMC interface, and button and touch-screen.
4. the embedded automatic pearl sorter based on image recognition according to claim 1, it is characterized in that: above described sorting executing agency, be provided with positioning strip, by arranging the state of infrared tube detection and location bar, obtained the position of pearl mechanism for sorting by the state reading infrared tube.
5. the embedded automatic pearl sorter based on image recognition according to claim 1, is characterized in that: described sorting executing agency also comprises the litter being provided with tooth bar, for coordinating the rotation of reducing motor.
6., based on an embedded automatic pearl method for sorting for image recognition, comprise the following steps:
(1) gather pearl image by image capture module, described image capture module adopts colour imagery shot;
(2) through core processing module, image is processed, judge the size of pearl, color, circularity, output sort control command;
(3) make pearl mechanism for sorting automatically classify to pearl according to the criteria for classification of setting, and the result data of classification is preserved.
7. the embedded automatic pearl method for sorting based on image recognition according to claim 1, is characterized in that: in step (2), image processing method is as described below,
1. the size of pearl is calculated by the ratio of the corresponding actual geometric area of the pixel of camera;
2. from image the valid pixel of pearl rgb value in obtain the color of pearl;
3. calculating pearl border center of gravity replaces the center of circle to calculate, and finds the minimum and maximum radius of target, judge the circularity of pearl with this by traversal.
8. the embedded automatic pearl method for sorting based on image recognition according to claim 7, it is characterized in that: step 1. in, described colour imagery shot carries out IMAQ to fixed size region, makes each pixel of camera and region area have fixing proportionate relationship.
9. the embedded automatic pearl method for sorting based on image recognition according to claim 8, it is characterized in that: step 1., 3. in calculate the size and shape of pearl, the coloured image of described pearl needs to carry out RGB successively and turns gray value, image filtering, image binaryzation process, edge detection, field, edge eight tracking step.
10. the embedded automatic pearl method for sorting based on image recognition according to claim 6, it is characterized in that: above pearl mechanism for sorting, be provided with positioning strip, and be provided with the infrared tube of detection and location bar state, by the position of infrared tube location pearl mechanism for sorting, the reducing motor controlling pearl mechanism for sorting moves left and right.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510617401.6A CN105195438B (en) | 2015-09-25 | 2015-09-25 | Embedded automatic pearl sorter and method for sorting based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510617401.6A CN105195438B (en) | 2015-09-25 | 2015-09-25 | Embedded automatic pearl sorter and method for sorting based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105195438A true CN105195438A (en) | 2015-12-30 |
| CN105195438B CN105195438B (en) | 2019-07-19 |
Family
ID=54943584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510617401.6A Active CN105195438B (en) | 2015-09-25 | 2015-09-25 | Embedded automatic pearl sorter and method for sorting based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105195438B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108719424A (en) * | 2018-06-04 | 2018-11-02 | 浙江海洋大学 | A kind of aquatic products sorting technique and system based on machine vision |
| CN110721927A (en) * | 2019-10-18 | 2020-01-24 | 珠海格力智能装备有限公司 | Visual inspection system, method and device based on embedded platform |
| CN112387606A (en) * | 2020-10-15 | 2021-02-23 | 西安工程大学 | Pearl sorting system |
| CN113308586A (en) * | 2021-06-09 | 2021-08-27 | 扬州哈工博视科技有限公司 | Material cutting and blanking auxiliary system for image recognition by screen flashing and working steps thereof |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101007308A (en) * | 2007-01-11 | 2007-08-01 | 浙江大学 | Pearl real time detection and classifying system based on mechanical vision |
| CN101149254A (en) * | 2007-11-12 | 2008-03-26 | 北京航空航天大学 | High accuracy vision detection system |
| CN201150919Y (en) * | 2008-01-17 | 2008-11-19 | 天津市华核科技有限公司 | Granular material sorting and grading device based on visual identification |
| WO2010040180A9 (en) * | 2008-10-09 | 2010-11-25 | Opal Producers Australia Limited | Modified apparatus and method for assessment, evaluation and grading of gemstones |
| JP2012047489A (en) * | 2010-08-24 | 2012-03-08 | K Mikimoto & Co Ltd | Method for determining nondestruction of pearl quality |
| CN102928431A (en) * | 2012-10-24 | 2013-02-13 | 浙江工业大学 | Device for automatically grading pearls on line according to size and shape on basis of monocular multi-view machine vision |
| CN103406289A (en) * | 2013-08-20 | 2013-11-27 | 浙江慧创科技有限公司 | Pearl feeding and image capturing device for sorting pearl luster |
| CN205074251U (en) * | 2015-09-25 | 2016-03-09 | 广东海洋大学 | Embedded pearl sorter based on image recognition |
-
2015
- 2015-09-25 CN CN201510617401.6A patent/CN105195438B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101007308A (en) * | 2007-01-11 | 2007-08-01 | 浙江大学 | Pearl real time detection and classifying system based on mechanical vision |
| CN101149254A (en) * | 2007-11-12 | 2008-03-26 | 北京航空航天大学 | High accuracy vision detection system |
| CN201150919Y (en) * | 2008-01-17 | 2008-11-19 | 天津市华核科技有限公司 | Granular material sorting and grading device based on visual identification |
| WO2010040180A9 (en) * | 2008-10-09 | 2010-11-25 | Opal Producers Australia Limited | Modified apparatus and method for assessment, evaluation and grading of gemstones |
| JP2012047489A (en) * | 2010-08-24 | 2012-03-08 | K Mikimoto & Co Ltd | Method for determining nondestruction of pearl quality |
| CN102928431A (en) * | 2012-10-24 | 2013-02-13 | 浙江工业大学 | Device for automatically grading pearls on line according to size and shape on basis of monocular multi-view machine vision |
| CN103406289A (en) * | 2013-08-20 | 2013-11-27 | 浙江慧创科技有限公司 | Pearl feeding and image capturing device for sorting pearl luster |
| CN205074251U (en) * | 2015-09-25 | 2016-03-09 | 广东海洋大学 | Embedded pearl sorter based on image recognition |
Non-Patent Citations (2)
| Title |
|---|
| 郑华文等: "基于计算机视觉的珍珠形状分级识别技术研究", 《工程设计学报》 * |
| 郑春煌等: "图像处理的珍珠形状大小检测系统研究", 《中国计量学院学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108719424A (en) * | 2018-06-04 | 2018-11-02 | 浙江海洋大学 | A kind of aquatic products sorting technique and system based on machine vision |
| CN110721927A (en) * | 2019-10-18 | 2020-01-24 | 珠海格力智能装备有限公司 | Visual inspection system, method and device based on embedded platform |
| CN112387606A (en) * | 2020-10-15 | 2021-02-23 | 西安工程大学 | Pearl sorting system |
| CN113308586A (en) * | 2021-06-09 | 2021-08-27 | 扬州哈工博视科技有限公司 | Material cutting and blanking auxiliary system for image recognition by screen flashing and working steps thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105195438B (en) | 2019-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103226088B (en) | Particulate counting method | |
| CN202021164U (en) | Peanut exterior quality detecting and sorting device | |
| CN102495069B (en) | Method for detecting defects of chain belts of zipper on basis of digital image processing | |
| CN103119449B (en) | Systems and methods to analyze an immunoassay test strip comb member | |
| CN105300854B (en) | Droplet parameter measuring apparatus and the droplet parameter measurement analysis method for utilizing the device | |
| CN105195438A (en) | Embedded type automatic pearl sorting device and method based on image recognition | |
| CN110044405B (en) | Automatic automobile instrument detection device and method based on machine vision | |
| CN106370667A (en) | Visual detection apparatus and method for quality of corn kernel | |
| CN107952696A (en) | A kind of detection grading plant and detection method suitable for fresh tobacco leaf | |
| CN104034638B (en) | The diamond wire online quality detecting method of granule based on machine vision | |
| CN1485616A (en) | Fowl eggs quality non-destruction automatic detection grading apparatus and process | |
| CN113421230B (en) | Visual detection method for defects of vehicle-mounted liquid crystal display light guide plate based on target detection network | |
| CN110533654A (en) | The method for detecting abnormality and device of components | |
| CN104777172A (en) | Quick and intelligent defective optical lens detection device and method | |
| CN205074251U (en) | Embedded pearl sorter based on image recognition | |
| CN105154988A (en) | Apparatus automatically extracting down feather and extracting method | |
| CN207222383U (en) | Plank sorting system | |
| CN107891012B (en) | Pearl size and circularity sorting device based on equivalent algorithm | |
| CN109509173A (en) | A kind of method of counting and counting device of bacterial clump | |
| CN114950969B (en) | Real-time visual identification and sorting system and method for main roots and stem bases of panax notoginseng | |
| CN102298134A (en) | Batch detection method and device for withdrawn electric energy meters | |
| CN103091330A (en) | Copper strip surface defect identifying device for simulating human visual perception mechanism and method thereof | |
| CN104034637A (en) | Diamond wire particle online quality inspection device based on machine vision | |
| CN115294024A (en) | Selenium drum defect detection method and device, electronic equipment and storage medium | |
| CN114378002A (en) | Hericium erinaceus grading system and grading method based on machine vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |