CN1329132C - On-line fruit quality detection and classification apparatus and method based on three camera chains - Google Patents
On-line fruit quality detection and classification apparatus and method based on three camera chains Download PDFInfo
- Publication number
- CN1329132C CN1329132C CNB2004100652162A CN200410065216A CN1329132C CN 1329132 C CN1329132 C CN 1329132C CN B2004100652162 A CNB2004100652162 A CN B2004100652162A CN 200410065216 A CN200410065216 A CN 200410065216A CN 1329132 C CN1329132 C CN 1329132C
- Authority
- CN
- China
- Prior art keywords
- fruit
- image
- camera
- speed
- classification
- 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.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/02—Food
- G01N33/025—Fruits or vegetables
Abstract
The present invention relates to a device and a method for detecting fruits. The device is composed of a fruit conveying and turning part, a computer vision identification part and a fruit grading and moving part. The present invention is characterized in that the computer vision identification part is composed of three progressive scanning color picture pick-up system, one light irradiating box, two photoelectric sensors, one synchronizer, a high-speed data communication module, a control module and corresponding software. Fruits which are conveyed at high speed freely rotate around a small horizontal axis to ensure that the fruits can be turned about 240 DEG when passing through the computer vision identification part; the computer vision identification part collects 9 pictures for each fruit through the three picture pick-up systems; quality and characteristic information is extracted from all the information of the pictures; then, identifying software is combined with a knowledge database to detect all the exterior qualities of the fruits, such as shapes, sizes, colors, defects, etc.; the fruit grading and moving part is controlled by a control module. The present invention can conveniently distinguish the fruit stems from the fruit calyxes of the fruits and can raise the precision of the detection of the shapes, the sizes and the defects of the fruits.
Description
Affiliated technical field
The present invention relates to a kind of characteristic and carry out the machine vision detection device and the method for real-time online detection and classification, refer in particular to a kind of method and apparatus that is used on the fruit high speed detection line based on three online fruit quality detection and classifications of camera system according to article.
Background technology
Along with the develop rapidly of computer technology, image processing techniques, mode identification technology, electronic technology, the classification that utilizes machine vision technique to carry out fruit obtains extensive studies.Abroad just begin one's study based on the fruit grading system of computer vision technique from early 1990s, the practical product of existing till now part, as the OSCARTM type and the MERLIN type high speed fruit grading production line of the U.S., be used for the grade deciding grade and level and the Quality Detection of fruit such as apple, pears, orange, peach.Related U.S. patent has, people's such as Yang Tao United States Patent (USP) " Method and apparatus for sorting objects by color (article being carried out the method and apparatus of classification by color) application number: 5339963 ", with a colour imagery shot article on the carrier chain are carried out fast detecting and classification, mainly use the colourity (H) in the HIS color system to carry out classified calculating.People's such as Yang Tao United States Patent (USP) " Method and apparatus for sorting objects by color including stable color tansformation (method and apparatus that article is carried out classification with color comprises a kind of conversion method of still image) application number: 5533628 " has been described a kind of fruit grading system based on single camera.Domestic, people such as Ying Yibin are arranged at a plurality of relevant Chinese patent (application number: also described a kind of fruit grading system 02136377.3,02266031.3,02160193.3 02295073.7,02295073.7) that they applied for based on single camera.These systems are made up of fruit conveying, turning system, Computer Vision Recognition system, hierarchy system.Fruit is fed forward with certain speed on conveying device, and fruit is freely rotated around trunnion axis, the image information of camera picked-up fruit surface.The grade of each fruit is comprehensively judged in Visual intelligent identification by computer vision system, and determines the positional information of each fruit, by the control module of computer recognition system instruction is transferred to hierarchy system, finishes the classification of fruit.
More than all open files all be to take to a fruit with single camera, though can reach requirement when some particular fruit is detected, but error ratio is bigger, mainly having following problem, 1) NI Vision Builder for Automated Inspection can only detect the surface of fruit about 90%: difference how to distinguish they and carpopodium and fruit calyx during 2) to defects detection; 3) because fruit mostly is non-regular object, when online detection, require fruit to be fed forward with certain speed, fruit is freely rotated around trunnion axis, fruit is difficult to be presented in the visual field of single camera with suitable attitude like this, brings error for like this testing result of shape and size; 4) used camera system mostly is general data capture card and camera composition, as the fruit quality of people such as Ying Yibin development detect in real time and grading robot system in camera system adopt be exactly the PanasonicCP410 colour TV camera that generates of the MatroxMe-teorII/Standard card produced of MATROX company and Toshiba form (" journal of Zhejiang university (agricultural and life science version) " the 30th rolled up for the 2nd phase, 2004), because captured fruit is high-speed motion, the fruit dynamic image of such system acquisition blurs, just image is a bit empty, they need to add that this step of image rectification carries out reverse and stick with paste computing in image is handled, can bring error to system like this.
The detection of fruit defects, be a global difficult problem, but be one that to detect in the sorted fruits classification, because if rotten fruit does not detect, fruit around it can cause in the process of transportation and storage all rots, and the economic loss that causes can be very big.Therefore many researchers are in this problem of research, solve people's such as the reasonable Yang of having Tao United States Patent (USP) " Item defect detection apparatus and method (a kind of method and apparatus of defects detection) " application number: 6.271520, this invention apparatus and method that two ir imaging systems come online detection fruit defects that on the fruit detection line, are arranged in parallel, one of them is the near-infrared camera system, another is middle ir imaging system, can easily carpopodium, state's calyx and defect area be branched away; With people's such as Yang Tao United States Patent (USP) " Defective object inspection and separation systemusing image analysis and curvature transformation (detect and distinguish the defective article " with image processing techniques and surface transfer) application number: 5732147, invented the easier on-line detection method of defective that a kind of volume imaging technology makes fruit.But these two kinds of methods all exist the detection cost too high, and it is accurate inadequately to detect data, and the problem of failing to solve carpopodium and the identification of fruit calyx.
Summary of the invention
For overcoming above-mentioned the deficiencies in the prior art, the invention provides a kind of apparatus and method based on three online fruit quality detection and classifications of camera system, can obtain those fruit external sort characteristic informations that high-speed motion also stirs uniformly on pipeline accurate and effective, all sidedly, can find carpopodium and fruit calyx apace, and finish exterior quality index ground such as fruit shape, size, color, defective on this basis and accurately detect, realize the real-time detection and the classification of fruit.
Technical scheme of the present invention is as follows:
Detection grading plant of the present invention is made up of fruit conveying, turning part, Computer Vision Recognition part and fruit grading action part.Computer Vision Recognition partly is installed on the fruit conveying, turning part, and the fruit grading action part is installed in the back of fruit conveying, turning part.
(application number: the fruit conveying, turning part 02266034.8) is similar for people's such as described fruit conveying, turning part and Ying Yibin utility model patent " fruit quality is intelligent to be detected and the classification production line ".The fruit conveying, turning device that the bipyramid roller that drives by chain advances promptly is set, and roller can freely rotate around the little axle of level, fruit must advance at utmost speed with 3～5/second speed, and guarantee by the Computer Vision Recognition part time, can rotate a week, the computer identification division is presented on whole surface.
Described fruit grading part can adopt utility model patent " fruit quality is intelligent to be detected and the classification the production line " (application number: the classification action part 02266034.8) of answering the refined people of grade of justice, also can adopt people's such as Zhang Cong utility model patent " a kind of fruit grader " (application number: the jet selecting device 03273524.3), require them can produce the classification action according to the graded signal that computer vision partly provides, the fruit of different stage is separated branch install in the different collecting boxs, finish fruit grading accurately.
Apparatus of the present invention are characterised in that: described Computer Vision Recognition part, it is core of the present invention, is made up of three color camera systems of lining by line scan, a lighting box, two photoelectric sensors, synchronized, high-speed data communications module, control module and a corresponding software.Wherein three colours of lining by line scan are taken the photograph in the system in each system and to be swept coloured silk line by line by one and retouch chromatic number word camera, a high-speed figure image pick-up card, a computer and be linked in sequence and form.Two photoelectric sensors are installed on the fruit conveying, turning device, link to each other with synchronized respectively by circuit, synchronized by circuit respectively with three computers in the high-speed figure image pick-up card link to each other with control module, the triggering signal of image taking and the position of each fruit are provided; Lighting box is installed in the top of fruit conveying, turning part, the inside is mounted with many fluorescent lamps, diffuse uniformly in the fruit surface generation, three colorful digital cameras of lining by line scan are installed in the lighting box, and in the same plane perpendicular to the fruit throughput direction, one of them camera is installed in the top of pipeline, other two both sides that are installed in pipeline, can photograph three fruit images of same position synchronously, three camera via line link to each other with high-speed figure image pick-up card in three computers respectively, deposit captured image in computer; Wherein a computer links to each other with control module by data wire, and control module is sent signal controlling corresponding classification executing agency and the classification driving mechanism is delivered to fruit in the corresponding collecting box.
Detection stage division of the present invention is characterized in that:
(1). the fruit conveying, turning partly makes fruit become single-row speed with 3～5/second to carry at a high speed and freely rotate around the little axle of level forward automatically with certain speed, guarantee by the Computer Vision Recognition part time, can overturn about 240 degree, the computer identification division is presented on the whole surface of fruit;
(2). the Computer Vision Recognition part is gathered 9 width of cloth images by being installed in perpendicular to three in the same plane of the fruit throughput direction colorful digital camera systems of lining by line scan to each fruit;
(3). the information that merges this 9 width of cloth image is finished all detections of exterior quality such as fruit shape, size, color, defective, comprehensively obtain the grade of each fruit, handle, and therefrom extract qualitative characteristics information, adopt Intelligent Recognition software to finish all detections of exterior quality such as fruit shape, size, color, defective again, and the classification instruction is sent to the fruit grading action part by control module in conjunction with the knowledge base that generates by existing standard and fruit database;
Particularly: 1). when a fruit in the process that advances at utmost speed while overturning, when reaching Computer Vision Recognition part first camera site, trigger three colorful digital camera systems of lining by line scan simultaneously by triggering sensor and synchronized, three colorful digital camera systems of lining by line scan are obtained the upper surface of two fruit of this fruit and front and the image of two sides, and rapidly the image of gained is carried out analyzing and processing, and when fruit reaches second camera site, finish image analysis processing;
2). when this fruit reaches second camera site, owing to advance at utmost speed while overturning during fruit, the fruit angle about 120 degree of having overturn, trigger sensor and synchronized once more and trigger three colorful digital camera systems of lining by line scan simultaneously, obtain image by above-mentioned job step, finish image analysis processing, and when fruit reaches the 3rd camera site, finish image analysis processing;
3). fruit reaches and acts as regent triple time when putting, and the fruit angle about 120 degree of overturning again and is obtained image by above-mentioned job step once more, and finished image analysis processing.
Described sequence image comprises each width of cloth Image Acquisition in handling, goes background, filtering, three fruit images are cut apart, the image of each fruit is carried out feature extraction respectively.
Described feature extraction comprises the extraction of fruit side information, the extraction of color characteristic, and on this basis to the carpopodium of this width of cloth fruit image, the really preliminary judgement of calyx, defective, shape, size.
Described high-speed data communications module requires three handled information of computer in time are fused in the computer rapidly, and it comprises the hardware and software that communication is used, must finish the communication of three intercomputers in the time of very short (Millisecond).
Adopt corresponding identification software three features that computer extracted in front and preliminary information judged to be merged in conjunction with the knowledge base of fruit grading standard and the generation of fruit database, finish all detections of exterior quality such as fruit shape, size, color, defective, obtain the grade of each fruit.
Light source in the lighting box can be selected and the position adjustment at different fruit objects, choose suitable light source frequency spectrum and locus, and make illumination uniformity in the visual field of three cameras, three positions of camera in light-source box also can be adjusted according to fruit object, make three cameras can obtain three fruit images.
During work, the bipyramid roller that fruit drives along with chain on the conveying, turning device on one side advances, under the effect of roller, freely rotate on one side around the little axle of level, three equidistant positions in the Computer Vision Recognition part along throughput direction, touch photoelectric sensor three times, each photoelectric sensor that triggers, photoelectric sensor sends to control module and computer by synchronized with the fruit position, and trigger three high-speed figure image pick-up cards in the computer simultaneously by synchronized, the image that collects fruit synchronously from three different directions, because three fruit images that three cameras can be simultaneously photograph same position from the top and the both sides of fruit, just fruit can be taken three times by same camera in the process of advancing, to obtain 9 width of cloth images to same fruit like this, the whole surface that guarantees fruit can both be arrived by the shooting system acquisition, three computers after having taken at every turn, in time captured sequence image is handled, and merge the information of 9 width of cloth images of each fruit by the high-speed data communications module, finish fruit shape by identification software, size, color, defectives etc. are the detection of exterior quality all, comprehensively obtains the grade of each fruit.By control module the classification instruction is sent to the automatic grading of fruits action part at last, finish fruit grading at last by the automatic grading of fruits action part.
Effect of the present invention is: (1) three camera systems of the present invention all adopt the color digital camera and the high-speed figure image pick-up card of lining by line scan, no matter how complicated the fruit motion is like this, how soon can both obtain fruit image clearly; (2) the present invention adopts the online detection of collection synchronously of three camera systems not relate in the file in the past, can obtain 9 width of cloth images to same fruit among the present invention, thereby the surface that guarantees fruit can both be collected, compare with one or two camera system, can not omit any information, if adopt three above camera systems, then redundancy is too much, increased the difficulty of back software data processing and communication, and processing speed is slack-off; (3) by the high-speed data communications module three resulting 9 width of cloth image informations of camera system are merged, can distinguish the carpopodium and fruit calyx of fruit very easily, thereby improve the shape of fruit and the accuracy of detection of size, and the precision that has improved defects detection; (4) the present invention gets up the information fusion in time of three camera systems by high-speed data communications module and process software at a high speed, guarantees that single line fruit grading speed is 3～5/second, can reach 3～4 tons/hour with regard to the apple sorting capability.
Description of drawings
Fig. 1 structural representation of the present invention;
Fig. 2 Computer Vision Recognition part-structure block diagram;
The same fruit f of Fig. 3
iThree are engraved in three width of cloth images that present in the same camera when different in lighting box;
Three cameras of Fig. 4 photograph three fruit images of same position from three different directions;
Fig. 5 Computer Vision Recognition part course of work schematic diagram;
The 1-lighting box, 2-fruit conveying, turning part, 3-synchronized, the 4-colorful digital camera a that lines by line scan, 5-high-speed figure image pick-up card a, 6-computer a, the 7-colorful digital camera b that lines by line scan, 8-high-speed figure image pick-up card b, 9-computer b, 10-high-speed data communications module, the 11-colorful digital camera c that lines by line scan, 12-high-speed figure image pick-up card c, 13-computer c, 14-control module, 15-automatic grading of fruits action part, the 16-photoelectric sensor, 17-light source, 18-fruit
The specific embodiment
As shown in Figure 1, the present invention is made up of fruit conveying, turning part 2, Computer Vision Recognition part and automatic grading of fruits action part 15, the structure of fruit conveying, turning part 2 and automatic grading of fruits action part 15 is with relevant patent " robot system of fruit quality detection in real time and classification " (application number: the Fig. 2 02136377.3), Fig. 4, shown in Figure 5, setting is advanced by the bipyramid roller that chain drives, and roller can be around little fruit conveying, turning device that freely rotates of level.Automatic grading of fruits action part 15 available devices are a lot, utility model patent " fruit quality is intelligent to be detected and the classification production line " (application number: the classification action part 02266034.8) as people such as Ying Yibin, utility model patent " a kind of fruit grader " (application number: the jet selecting device 03273524.3) with people such as Zhang Cong, select the former for use in this patent, promptly application number is the classification action part in 02266034.8.
As shown in Figure 1 and Figure 2, the Computer Vision Recognition part: form 14 by three color camera systems of lining by line scan, lighting box 1, two photoelectric sensors 16, synchronized 3, high-speed data communications module 10, a control module, wherein first colour of lining by line scan is taken the photograph system and is linked in sequence and is formed by the colorful digital camera 4a that lines by line scan, high-speed figure image pick-up card 5a, computer 6a.Second colour of lining by line scan taken the photograph system and is linked in sequence and forms for the colorful digital camera 7b that lines by line scan, high-speed figure image pick-up card 8b, computer 9b.The 3rd colour of lining by line scan taken the photograph system and is linked in sequence and forms for the colorful digital camera 11c that lines by line scan, high-speed figure image pick-up card 12c, computer 13c.Two photoelectric sensors 16 are installed on the fruit conveying, turning part 2, link to each other with synchronized 3 respectively by circuit, synchronized 3 links to each other with high-speed figure image pick-up card 5a, 8b, 12c respectively by circuit, and link to each other with control module 14, the triggering signal of image taking and the position of each fruit 18 are provided; Lighting box 1 is installed in the top of fruit conveying, turning part 2, the inside is mounted with many fluorescent lights 17, diffuse uniformly in fruit 18 surface generations, three colorful digital camera 4a that line by line scan in lighting box 1,7b, in 11c is installed in plane perpendicular to the fruit throughput direction from top and both sides respectively, and guarantee at these three camera 4a, 7b, 11c can photograph three fruit images of same position synchronously from both sides and top, three camera 4a, 7b, the 11c via line respectively with high-speed figure image pick-up card 5a, 8b, 12c, high-speed figure image pick-up card 5a, 8b, the 12c 6a that links to each other with three computers respectively, 9b, 13c; Link to each other by high-speed data communications module 10 between three computer 6a, 9b, 13c, merge captured information at any time; Wherein a computer 13c links to each other with control module 14 by data wire, and control module 14 is sent signal and fruit delivered in the corresponding collecting box for automatic grading of fruits action part 15 corresponding classification executing agencies of control and classification driving mechanism.
According to the fruit that is detected, select the light source 18 of suitable frequency spectrum to be installed in the lighting box 1, and the position of definite light source 18, make the illumination uniformity in the visual field of three cameras.Adjust three 4a, 7b, the position of 11c camera in light-source box simultaneously, make each camera can obtain the image of a plurality of fruit.
As shown in Figure 5, during work, the bipyramid roller that fruit drives along with chain on conveying, turning device 2 on one side advances, under the effect of roller, freely rotate on one side around the little axle of level, during camera site in reaching lighting box 1, touch photoelectric sensor 16, photoelectric sensor 16 sends to control module 14 and computer 13 by synchronized 3 with the fruit position, and triggering high-speed figure image pick-up card 5a, 8b, 12c simultaneously by synchronized 3, high-speed figure image pick-up card 5a, 8b, 12c import fruit image in three computers into.The image that collects fruit synchronously from three different directions, because three colorful digital camera 4a that line by line scan, 7b, 11c can photograph three fruit images of same position as shown in Figure 4 simultaneously from three different directions, just fruit 18 can be taken three times by same camera in the process of advancing, to obtain nine width of cloth images to same fruit like this, the whole surface that guarantees fruit 18 can both be arrived by the shooting system acquisition, three computer 6a, 9b, 13c handles by the information that high-speed data communications 10 modules merge this nine width of cloth image, and therefrom extract qualitative characteristics information, finish fruit shape by Intelligent Recognition software in conjunction with the knowledge base that generates by existing standard and fruit database, size, color, defectives etc. are the detection of exterior quality all, comprehensively obtain the grade of each fruit, and classification instruction is sent to automatic grading of fruits action part 15 by control module 14, finish fruit grading at last by automatic grading of fruits action part 15.
Described fruit 18 is taken three times by same camera in the process of advancing be to realize like this, as shown in Figure 3, fruit is rotating and is passing through three interior positions of lighting box interior image that is presented of camera therein, and fruit fi is in three positions of 3 moment t0 to t2 among Fig. 3.Like this, just can obtain 3 width of cloth images to each fruit from same camera, then three cameras can obtain 9 width of cloth images, thereby the information that guarantees the fruit all surface can collect.
Claims (6)
1. based on three online fruit quality detection and classification devices of camera system, form by fruit conveying, turning part, Computer Vision Recognition part and fruit grading action part, Computer Vision Recognition partly is installed on the fruit conveying, turning part, and the fruit grading action part is installed in the back of fruit conveying, turning part; The fruit conveying, turning device that described fruit conveying, turning partly advances for the bipyramid roller that drives by chain, and roller can freely rotate around the little axle of level; Described fruit grading partly adopts the graded signal that can partly provide according to computer vision to produce the classification action part or the jet selecting device be made up of classification executing agency and classification driving mechanism of classification action;
It is characterized in that: described Computer Vision Recognition part is by three color camera systems of lining by line scan, a lighting box (1), two photoelectric sensors (16), a synchronized (3), high-speed data communications module (10), control module (14) and corresponding software are formed, wherein three colours of lining by line scan are taken the photograph and are swept coloured silk line by line by one in each system in the system and retouch chromatic number word camera (7), a high-speed figure image pick-up card (5), a computer (6) is linked in sequence and forms, two photoelectric sensors (16) are installed on the fruit conveying, turning device (2), link to each other with synchronized (3) respectively by circuit, synchronized (3) by circuit respectively with three computer (6a, 9b, high-speed figure image pick-up card (5a 13c), 8b, 12c) link to each other with control module (14); Lighting box (1) is installed in the top of fruit conveying, turning part (2), the inside is mounted with the light source (17) that many fluorescent lamps are formed, three colorful digital camera (4a that line by line scan, 7b, 11c) be installed in the lighting box (1), and in the same plane perpendicular to the fruit throughput direction, one of them camera (4a) is installed in the top of pipeline, other two (7b, 11c) be installed in the both sides of pipeline, three camera (4a, 7b, 11c) via line respectively with three computer (6a, 9b, 13c) Nei high-speed figure image pick-up card (5a, 8b, 12c) link to each other; Wherein a computer (12c) links to each other with control module (14) by data wire, and control module (14) links to each other with classification driving mechanism (15) with corresponding classification executing agency.
2. based on three the online fruit quality detection and classification Device Testing of camera system methods, it is characterized in that:
(1). the fruit conveying, turning partly makes fruit become single-row speed with 3～5/second to carry at a high speed and freely rotate around the little axle of level forward automatically with certain speed;
(2). the Computer Vision Recognition part is gathered 9 width of cloth images by being installed in perpendicular to three in the same plane of the fruit throughput direction colorful digital camera systems of lining by line scan to each fruit;
(3). the information that merges this 9 width of cloth image is finished all detections of exterior quality such as fruit shape, size, color, defective, comprehensively obtain the grade of each fruit, handle, and therefrom extract qualitative characteristics information, adopt Intelligent Recognition software to finish all detections of exterior quality such as fruit shape, size, color, defective again, and the classification instruction is sent to the fruit grading action part by control module in conjunction with the knowledge base that generates by existing standard and fruit database.
3. detection method according to claim 2 is characterized in that:
(1). when a fruit in the process that advances at utmost speed while overturning, when reaching Computer Vision Recognition part first camera site, trigger three colorful digital camera systems of lining by line scan simultaneously by triggering sensor and synchronized, three colorful digital camera systems of lining by line scan are obtained the upper surface of two fruit of this fruit and front and the image of two sides, and rapidly the image of gained is carried out analyzing and processing, when reaching second camera site, handle fruit;
(2). when this fruit reaches second camera site, owing to advance at utmost speed while overturning during fruit, the fruit angle about 120 degree of having overturn, trigger sensor and synchronized once more and trigger three colorful digital camera systems of lining by line scan simultaneously, obtain image by above-mentioned job step, finish image analysis processing, and when fruit reaches the 3rd camera site, finish image analysis processing;
(3). fruit reaches and acts as regent triple time when putting, and the fruit angle about 120 degree of overturning again and is obtained image by above-mentioned job step once more, and finished image analysis processing.
4. detection method according to claim 2 is characterized in that: described sequence image comprises each width of cloth Image Acquisition in handling, goes background, filtering, three fruit images are cut apart, the image of each fruit is carried out feature extraction respectively.
5. detection method according to claim 2, it is characterized in that: described feature extraction comprises the extraction of fruit side information, the extraction of color characteristic, and on this basis to the preliminary judgement of the carpopodium of this width of cloth fruit image fruit calyx, defective, shape, size.
6. detection method according to claim 2, it is characterized in that: described high-speed data communications module comprises the hardware and software that communication is used, three handled information of computer in time are fused in the computer rapidly, in the time of Millisecond, finish the communication of three intercomputers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100652162A CN1329132C (en) | 2004-11-02 | 2004-11-02 | On-line fruit quality detection and classification apparatus and method based on three camera chains |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100652162A CN1329132C (en) | 2004-11-02 | 2004-11-02 | On-line fruit quality detection and classification apparatus and method based on three camera chains |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1603013A CN1603013A (en) | 2005-04-06 |
| CN1329132C true CN1329132C (en) | 2007-08-01 |
Family
ID=34666484
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100652162A Expired - Fee Related CN1329132C (en) | 2004-11-02 | 2004-11-02 | On-line fruit quality detection and classification apparatus and method based on three camera chains |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1329132C (en) |
Families Citing this family (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100534642C (en) * | 2007-09-24 | 2009-09-02 | 浙江大学 | Fruit classifying method according to surface color |
| BE1018303A3 (en) * | 2008-10-10 | 2010-08-03 | Bestray | METHOD FOR DIVIDING A STONE FRUIT IN A PRELIMINARY CLASS. |
| CN101486033B (en) * | 2009-03-09 | 2012-07-25 | 中国农业大学 | Device for sorting fruits and method for sorting Chinese chestnut |
| CN101957196A (en) * | 2010-07-15 | 2011-01-26 | 上海大学 | Machine vision based automatic detection device of anastomat |
| CN101984346A (en) * | 2010-10-19 | 2011-03-09 | 浙江大学 | Method of detecting fruit surface defect based on low pass filter |
| CN102506743A (en) * | 2011-11-15 | 2012-06-20 | 中山市德唯纳电子科技有限公司 | Automatic corrugated pipe detection equipment and detection method |
| CN103185609A (en) * | 2011-12-29 | 2013-07-03 | 机械科学研究总院先进制造技术研究中心 | Image detecting method for grading of tomatoes |
| CN103240227A (en) * | 2013-05-17 | 2013-08-14 | 苏州福丰科技有限公司 | Saw blade sorting device capable of distinguishing by images |
| JP6379454B2 (en) * | 2013-07-31 | 2018-08-29 | シンフォニアテクノロジー株式会社 | Image processing apparatus for parts feeder and parts feeder |
| CN103406286B (en) * | 2013-08-16 | 2015-06-24 | 常熟理工学院 | Online grading method for external quality of fruit based on LabVIEW |
| CN103487147B (en) * | 2013-09-06 | 2016-03-23 | 闻泰通讯股份有限公司 | Electronic equipment is utilized to select the system and method for fruit |
| CN105344620B (en) * | 2015-10-14 | 2018-10-23 | 合肥安晶龙电子股份有限公司 | Color selection method based on material shapes |
| CN105597911A (en) * | 2016-01-07 | 2016-05-25 | 何天柱 | Fruit sorting method |
| CN105806743A (en) * | 2016-04-28 | 2016-07-27 | 西北农林科技大学 | Multi-view apple moldy core detection device and method |
| CN105834119A (en) * | 2016-05-04 | 2016-08-10 | 成都贝森伟任科技有限责任公司 | Automatic fruit grading device |
| CN107520144A (en) * | 2016-06-20 | 2017-12-29 | 新疆农业大学 | A kind of "Hami" melon grader based on machine vision technique |
| CN107413674B (en) * | 2017-06-08 | 2019-09-10 | 北京农业信息技术研究中心 | A kind of crop seed on-line sorting method and system |
| CN107694962A (en) * | 2017-11-07 | 2018-02-16 | 陕西科技大学 | A kind of fruit automatic sorting method based on machine vision and BP neural network |
| CN107931153A (en) * | 2018-01-10 | 2018-04-20 | 陈德言 | A kind of poor quality tablet automatic separating apparatus |
| CN109827971B (en) * | 2019-03-19 | 2021-09-24 | 湖州灵粮生态农业有限公司 | Method for nondestructive detection of fruit surface defects |
| CN110813796A (en) * | 2019-11-19 | 2020-02-21 | 龙口市汇源果蔬有限公司 | Apple cleaning and sorting system and process |
| CN112808603B (en) * | 2020-12-22 | 2021-10-12 | 南京林业大学 | Fresh cut flower sorting device and method based on RealSense camera |
| CN112958491A (en) * | 2021-03-31 | 2021-06-15 | 安徽理工大学 | Part vision detection system based on DSP and FPGA |
| CN112974303B (en) * | 2021-04-26 | 2022-11-08 | 山东深蓝智谱数字科技有限公司 | Hyperspectrum-based fruit quality detection method, device and medium |
| CN113680706B (en) * | 2021-05-19 | 2022-12-23 | 苏州大学 | Pneumatic separation coal gangue sorting method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5917926A (en) * | 1996-03-01 | 1999-06-29 | Durand-Wayland, Inc. | Optical inspection apparatus and method for articles such as fruit and the like |
| US6271520B1 (en) * | 1998-03-23 | 2001-08-07 | University Of Arkansas | Item defect detection apparatus and method |
| CN2555110Y (en) * | 2002-08-03 | 2003-06-11 | 浙江大学 | Intelligence detection of fruit quality and its sorting prodn line |
| CN2635228Y (en) * | 2003-08-18 | 2004-08-25 | 广东省农业机械研究所 | Fruits grading machine |
-
2004
- 2004-11-02 CN CNB2004100652162A patent/CN1329132C/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5917926A (en) * | 1996-03-01 | 1999-06-29 | Durand-Wayland, Inc. | Optical inspection apparatus and method for articles such as fruit and the like |
| US6271520B1 (en) * | 1998-03-23 | 2001-08-07 | University Of Arkansas | Item defect detection apparatus and method |
| CN2555110Y (en) * | 2002-08-03 | 2003-06-11 | 浙江大学 | Intelligence detection of fruit quality and its sorting prodn line |
| CN2635228Y (en) * | 2003-08-18 | 2004-08-25 | 广东省农业机械研究所 | Fruits grading machine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1603013A (en) | 2005-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1329132C (en) | On-line fruit quality detection and classification apparatus and method based on three camera chains | |
| CN109724984B (en) | Defect detection and identification device and method based on deep learning algorithm | |
| CN106238342B (en) | Panoramic vision potato sorts and defect detecting device and its sorting detection method | |
| CN103323457B (en) | Fruit appearance defect real-time online detection system and detection method | |
| CN102495069B (en) | Method for detecting defects of chain belts of zipper on basis of digital image processing | |
| CN101627832B (en) | On-line automatic tobacco recognition and classification system | |
| CN206139527U (en) | Panoramic vision potato is selected separately and defect detecting device | |
| CN105388162B (en) | Raw material silicon chip surface scratch detection method based on machine vision | |
| CN100547394C (en) | Fruit quality detection system based on image information fusion technology | |
| CN102621156B (en) | Image-processing-based automatic micro part sorting system | |
| CN106770331A (en) | A kind of workpiece, defect detecting system based on machine vision | |
| CN107966454A (en) | A kind of end plug defect detecting device and detection method based on FPGA | |
| CN104483320B (en) | Digitized defect detection device and detection method of industrial denitration catalyst | |
| CN202974886U (en) | Machine-vision-based tank body seam defect detection system | |
| CN105651782A (en) | Engine cylinder cover defect automatic detection equipment based on machine vision | |
| CN206022316U (en) | A kind of solar energy silicon crystal chip defect automatic checkout equipment | |
| CN107084992A (en) | A kind of capsule detection method and system based on machine vision | |
| CN109030504B (en) | Panoramic potato defect detection device and method based on virtual and real imaging combination | |
| CN206057205U (en) | A kind of notebook computer assembly quality automatic detection device | |
| CN210071686U (en) | Fruit grading plant based on orthogonal binocular machine vision | |
| CN203245133U (en) | Lens defect detecting system | |
| CN110146516A (en) | Fruit sorter based on orthogonal binocular machine vision | |
| CN110044921A (en) | Lithium battery open defect detection system and method | |
| CN106680293A (en) | Machine vision based motor rotor gib defect detection system and method | |
| CN106778439A (en) | A kind of lithium battery batch scan code system and method based on image procossing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C19 | Lapse of patent right due to non-payment of the annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |