CN102773217B - Automatic grading system for kiwi fruits - Google Patents
Automatic grading system for kiwi fruits Download PDFInfo
- Publication number
- CN102773217B CN102773217B CN 201210296845 CN201210296845A CN102773217B CN 102773217 B CN102773217 B CN 102773217B CN 201210296845 CN201210296845 CN 201210296845 CN 201210296845 A CN201210296845 A CN 201210296845A CN 102773217 B CN102773217 B CN 102773217B
- Authority
- CN
- China
- Prior art keywords
- kiwi berry
- image
- controller
- camera
- processor
- 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
Landscapes
- Sorting Of Articles (AREA)
Abstract
The invention discloses an automatic grading system for kiwi fruits. The automatic grading system comprises a single-row slide plate device, a single positioning transmission device, a first electromotor, a first position sensor, a transparent conveyor belt, a second electromotor, a second position sensor, a first camera, a second camera, a grading executive system, a processor and a controller, wherein the kiwi fruits are arrayed in a single row by the single-row slide plate device and the single positioning transmission device and a certain distance between every two kiwi fruits is maintained; the first electromotor and the second electromotor are controlled by the controller to ensure that the single kiwi fruits can be static in an image acquisition area; the kiwi fruits static on the transparent conveyor belt can be subjected to image acquisition by the first camera and the second camera; the all-around clear images of the kiwi fruits can be acquired; and after the image characteristic of the acquired image is extracted by the processor, damage detection on the acquired image is performed by back propagation (BP) network, and correct grading results are output by combining thecharacteristics of the sizes and the shapes of the kiwi fruits.
Description
Technical field
The present invention relates to the classification technique field of Kiwi berry, more particularly, relate to the automatic hierarchy system of a kind of Kiwi berry.
Background technology
Kiwi berry is the abundant fruit of a kind of nutritive value, has multi-efficiency and effect, is called king in the fruit by people, and its market price is considerable, has very big consumption potentiality in domestic and international market.Statistics according to fruit industry office shows, by the end of the year 2010, and 2,000,000 mu of global Kiwi berry cultivated areas, China accounts for 1,000,000 mu, but Kiwi berry classification in postpartum is handled ratio less than 2%, and developed country handles ratio postpartum and accounts for 80%～90%.Therefore, China's Kiwi berry is lacked competitiveness in the international market.
The existing Kiwi berry classification technique of China mainly by gathering the image of the Kiwi berry in the tumbling motion, is exported classification results to the image that collects then after treatment.Because existing classification technique collection is image in the Kiwi berry tumbling motion, therefore, control to Kiwi berry is comparatively complicated, and the image of the Kiwi berry that collects is more complicated and fuzzy also, when the image that collects is handled, need deblurring and image complicated processing such as to cut apart, make to the classification of Kiwi berry consuming time long and classification results is inaccurate.
Summary of the invention
In view of this, the invention provides the automatic hierarchy system of a kind of Kiwi berry, to realize that Kiwi berry is carried out convenient, fast and accurate classification.
For solving the problems of the technologies described above, the technical solution used in the present invention is: the automatic hierarchy system of a kind of Kiwi berry comprises: single file skate apparatus, single location conveying device, first motor, primary importance sensor, transparent transmission band, second motor, second place sensor, first camera, second camera, processor, controller and classification executive system; Wherein:
Described single file skate apparatus links to each other with described single location conveying device with storage device respectively, be used for to regulate the conveying capacity of the Kiwi berry that falls from storage device, makes it to become single-row shape and is delivered in the conveying device of described single location;
Described first motor links to each other with single location conveying device with described controller respectively, and the control signal that is used for sending according to described controller when classification drives described single location conveying device and carries Kiwi berry;
Described primary importance sensor is positioned at the conveying end of described single location conveying device, and link to each other with described controller, be used for when Kiwi berry arrives described single location delivery device tip, gathering the information that described Kiwi berry arrives, and the information that the described Kiwi berry that will gather arrives is sent to controller;
Described controller is controlled described first motor according to the information that receives and is shut down;
Described transparent conveyer belt links to each other with second motor with described single location conveying device respectively, is used for transmitting falling to the Kiwi berry of described transparent conveyer belt to image acquisition region from described single location conveying device;
Described second sensor is positioned at image acquisition region and links to each other with described controller, be used for when Kiwi berry arrives image acquisition region with described transparent conveyer belt, gathering described Kiwi berry information in place, and the described Kiwi berry information in place that will gather is sent to controller;
Described controller shuts down according to the second coupled motor of information control that receives, and the information that this Kiwi berry is in place sends to processor;
Described processor adopting triggers effective means, present frame Kiwi berry image in all Kiwi berry images of first camera in the image acquisition region and second camera collection is as effective Kiwi berry image, and will described effective Kiwi berry image be sent to image pre-processing module in the processor as the processing object; Described first camera and second camera are gathered the described Kiwi berry image of two sides up and down respectively;
Described processor receives and handles described effective Kiwi berry image, according to result output classification results;
Described controller also links to each other with described processor, be used for to receive the classification results of described processor output, and controls described classification executive system according to described classification results Kiwi berry is dispensed to sorter frame;
Wherein, described processor comprises: human-computer interaction module, image pre-processing module, characteristic extracting module and feature Fusion Module; Wherein:
Described human-computer interaction module is by sending control command to controller, realize the controlling and monitoring of equipment, and coordinate image processing, IMAQ, controller three and work in order;
Described image pre-processing module is cut apart with the image cutting by image denoising, image the image of described first camera and second camera collection is carried out preliminary treatment; Described image denoising adopts small echo denoising mode;
Described characteristic extracting module is extracted size, shape and blemish feature from the pretreated image of process;
Described feature Fusion Module merges described size, shape and the blemish feature of extracting, and obtains classification results according to fusion results.
Preferably, described first camera links to each other with described processor by USB interface respectively with second camera.
Preferably, described controller is the STC89C52 single-chip microcomputer.
Preferably, also comprise the classification executive system that links to each other with described controller;
Described controller also links to each other with described processor, be used for to receive the classification results of described processor output, and controls described classification executive system according to described classification results Kiwi berry is dispensed to sorter frame.
From above-mentioned technical scheme as can be seen, the automatic hierarchy system of a kind of Kiwi berry disclosed by the invention, by single file skate apparatus and single location conveying device, can make Kiwi berry be arranged as the single file shape and can keep the distance of 4-6cm between any two, when the primary importance sensor detects Kiwi berry and arrives single location conveying device terminal, first motor that controller control drives single location conveyor running stops, can guarantee in a period of time, to have only a Kiwi berry on transparent conveyer belt, to move to image acquisition region, when second place sensor detects Kiwi berry arrival image acquisition region, second motor that controller control drives transparent conveyer belt running stops, can make the single image acquisition region that is still in of Kiwi berry, first camera and second camera carry out IMAQ up and down to the Kiwi berry that is still on the transparent transmission band, can make Kiwi berry need not the rolling running and just can collect comparatively comprehensively image of Kiwi berry, by processor classification results is handled and exported to the Kiwi berry image of gathering, controller restarts first motor and next Kiwi berry is carried out classification handles after a last Kiwi berry classification processing finishes and enters corresponding sorter frame.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the structural representation of the automatic hierarchy system of the disclosed a kind of Kiwi berry of the embodiment of the invention.
The specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is a part of embodiment of the present invention, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that obtains under the creative work prerequisite.
The embodiment of the invention discloses the automatic hierarchy system of a kind of Kiwi berry, to realize that Kiwi berry is carried out convenient, fast and accurate classification.
As shown in Figure 1, make hierarchy system by oneself for the disclosed a kind of Kiwi berry of the embodiment of the invention, comprising:
Single file skate apparatus 1, single location conveying device 2, first motor 3, primary importance sensor 4, transparent transmission band 5, second motor 6, second place sensor 7, first camera 8, second camera 9, processor 10, controller 11 and classification executive system 12; Wherein:
Single file skate apparatus 1 links to each other with single location conveying device 2 with storage device respectively, be used for to regulate the conveying capacity of the Kiwi berry that falls from storage device, makes it to become single-row shape and is delivered in the single location conveying device 2;
First motor 3 links to each other with single location conveying device 2 with controller 11 respectively, and when being used for starting working in system, the control signal of sending according to controller 11 drives single location conveying device 2 and carries Kiwi berrys;
First camera 8 and second camera 9 also link to each other by USB interface with processor 10 respectively, and the image that is used for gathering is sent to processor 10;
The classification results of controller 11 receiving processors 10 output, and according to described classification results control classification executive system 12 Kiwi berry is dispensed to sorter frame.
After the classification results and (current Kiwi berry enters corresponding sorter frame) of processor 10 output Kiwi berrys, controller 11 sends control signal, controls first motor 3 and restarts running, and next Kiwi berry is transmitted, so circulation is carried out classification to all Kiwi berrys.
In above-mentioned, controller 11 is the STC89C52 single-chip microcomputer.
Characteristic extracting module is extracted size, shape and blemish feature from the pretreated image of process;
The feature Fusion Module merges described size, shape and the blemish feature of extracting, and obtains classification results according to fusion results.
Concrete, because the image of camera collection comprises information such as background, noise, need at first image to be carried out preliminary treatment, and the image preliminary treatment is the committed step that image is handled graphical analysis, can be follow-up feature extraction, image recognition work provides Useful Information;
In the preprocessing process of image, adopting medium filtering, mean filter, adaptive wiener filter and small echo denoising that the gray level image of Kiwi berry is carried out filtering respectively handles, image after the contrast denoising is found: strong through the picture contrast after the small echo denoising, fuzzy few, the definition height, image border and detail section information are preserved more complete;
To carrying out rim detection through the image after the small echo denoising, find real more border;
The image that rim detection is obtained adopts bigger expansion parameter to expand, then to the image after expanding fill, operations such as burn into opening operation, deletion redundant sub-pixels, finally obtain the Kiwi berry area image;
With the Kiwi berry area image as template, respectively with the small echo denoising after R, G, B component carry out and computing the colored Kiwi berry image that obtains from background, splitting finally by compose operation;
For the ease of follow-up Surface Defect Recognition and Shape Feature Extraction, a large amount of white background is dismissed in the image after image need being cut apart, four extreme position coordinates of searching image upper and lower, left and right with this coordinate information input cutting function, obtain images cut;
After the image preliminary treatment, from pretreated image, extract size, shape and blemish feature; Concrete, processor is chosen based on the match relation of area image pixel extraction, actual weight and amount of pixels and threshold value the Kiwi berry size characteristic is extracted;
The area image amount of pixels is extracted:
Because camera is taken the Kiwi berry image of two sides up and down respectively, and the fixed-site of camera, focal length mixes up in advance, and all shooting environmental immobilize, so can select for use Kiwi berry area image amount of pixels to represent its bulk very easily.Consider that lopsided Kiwi berry there are differences on two side sizes up and down, two width of cloth images up and down that camera is taken extract the area image amount of pixels simultaneously, are averaged, as the judgment basis of Kiwi berry size;
The match relation of actual weight and amount of pixels:
Rule of 10 not of uniform size and shapes is chosen in experiment, with kind with batch Kiwi berry as sample, extract its mean pixel information of two sides up and down respectively, and take by weighing its actual weight, adopt the two relation of least square fitting, draw the Kiwi berry mean pixel amount Y(unit of two sides up and down: individual) weight W actual with it (unit: g) satisfy linear approximate relationship as follows:
Y=798W+34081(1)；
Threshold value is chosen:
According to the agricultural industry criteria NY/T1794-2009 of the People's Republic of China (PRC) of issue on December 22nd, 2009, Kiwi berry size specification standard and convolution (1) draw each specification amount of pixels threshold value of Kiwi berry.
Processor extracts based on Fourier transform pairs Kiwi berry shape facility:
Concrete, by the border follow the tracks of obtain border sequence select, obtain center point coordinate, calculate radius sequence and normalization, to the radius sequence make discrete Fourier transform, normalization Fourier descriptor, definition out-of-shape degree is described and the Kiwi berry Shape Feature Extraction is carried out in concrete experiment.
Processor is identified the blemish feature based on the BP network:
Concrete, at the common scab symptom of Kiwi berry, with the actual scab of blackspot simulation Kiwi berry, obtain the image approximate component by wavelet decomposition, and extracting its statistical nature for BP network, the BP network after the training is realized the Surface Defect Recognition of Kiwi berry as grader.
Wavelet decomposition:
Wavelet transformation is a kind of new transform analysis method, and its main feature is by the abundant feature of some aspect of outstanding problem of conversion, is widely used in Digital Image Processing.The wavelet transformation W of arbitrary signal f (t)
f(a b) is defined as:
In the formula (2): ψ (t) is female small echo, and a is contraction-expansion factor, and b is shift factor.This shows that wavelet transformation is the signal of binary form, and wavelet transformation inhibit signal inner product is constant, it is at function space L
2(R) have complete inverse transformation relation on, wavelet transformation is non-loss transformation.
Every layer of wavelet decomposition all resolves into a low frequency part and three HFSs, carries out subordinate when decomposing, and low frequency part is continued to decompose, and low frequency part namely is the thumbnail of original image, and HFS has kept detailed information such as the texture, edge of image.Through repetition test, choose 2 rank Daubechies wavelet basis to carrying out the defect area feature that three layers of wavelet decomposition can embody Kiwi berry best through pretreated Kiwi berry gray level image.The image that the image preliminary treatment is obtained is converted into gray level image, and this gray level image is carried out three layers of wavelet decomposition with 2 rank Daubechies wavelet basis.
The BP network classification:
The defect area feature is the most obvious, and the approximate component of the minimum ground floor of interference such as edge is compressed into the matrixes of 100 * 100 dimensions, and extracts the statistical nature of this matrix: intermediate value, maximum, minimum of a value, maximum and minimum of a value poor.Choose 100 of representative training samples from a large amount of Kiwi berry images of taking in advance, input BP network is trained after extracting above-mentioned 4 statistical natures and adopting the extreme value method for normalizing to carry out normalized.To above-mentioned 4 statistical natures of the same extraction of the Kiwi berry image of taking in the equipment actual operation, utilize the maximum of training sample matrix, minimum of a value that it is made normalized, the BP network after the input training is finally realized the Kiwi berry Surface Defect Recognition.
Merge described size, shape and the blemish feature extracted:
Concrete, according to the Kiwi berry agricultural industry criteria and in conjunction with consumer's hobby, fusion is at size characteristic that feature extraction phases is extracted: big (L), in (M), little (S), shape facility: good, poor and surface characteristics: surperficial defectiveness, zero defect feature, Kiwi berry is divided into four grades the most at last, has the Kiwi berry of scab defective to be divided into fourth class product with every.
In the above-described embodiments, control first motor and second motor by controller, make that Kiwi berry can be by single location conveying device and the single image acquisition region that is still in of transparent transmission band, solved and gathered the Kiwi berry image in the motion in traditional Kiwi berry hierarchy system and cause image blurring, programming complicated problems; And by two cameras that lay respectively at transparent conveyer belt upper and lower Kiwi berry is carried out IMAQ in the above-described embodiments, make that the Kiwi berry image that collects is comparatively complete.
Concrete, in the automatic classification process to Kiwi berry, at first, utilize human-computer interaction interface on the processor 10 that classifying equipoment (man-machine interface mainly provide startups, time-out information such as parameter in continuation, shut-down operation and Kiwi berry classification results, classification progress, the classification) is provided, realize connecting between the initialization of controller 11 and it and the processor 10 simultaneously, subsequently, controller 11 determines first motor 3 of single location conveying device 2 whether to shut down according to whether receiving the Kiwi berry arriving signal that primary importance sensor 4 sends; If controller 11 receives the Kiwi berry arriving signal that primary importance sensor 4 sends, then controlling single location conveying device 2 stops, (Kiwi berry is after dropping on the conveyer belt, passed through horizontal positioning device, make each Kiwi berry at the attitude basically identical of image acquisition areas) monitor the Kiwi berry signal in place that second place sensor 7 transmits when controller 11, and whether shut down according to second motor 6 of the transparent conveyer belt 5 of this signal deciding classification; If controller 11 receives the Kiwi berry signal in place that second place sensor 7 sends, then control transparent conveyer belt 5 stop motions of classification, controller 11 sends Kiwi berry signal in place by serial communication to processor 10; First camera 8 and 9 pairs of Kiwi berrys of second camera carry out IMAQ subsequently, and read the Kiwi berry image of present frame by the image processing program of processor 10, and image handled, obtain the quality of Chinese gooseberries characteristic parameter, the network based quality characteristic parameter of BP provides the Kiwi berry rank subsequently, and by serial communication graded signal is reached controller 11 by processor 10; Subsequently, controller 11 moves according to corresponding classification baffle plate in the graded signal control classification executive system 12, and restarts transparent conveyer belt 5 motions of second motor, 6 drive classifications, realizes the classification of Kiwi berry.
Each embodiment adopts the mode of going forward one by one to describe in this specification, and what each embodiment stressed is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.
To the above-mentioned explanation of the disclosed embodiments, make this area professional and technical personnel can realize or use the present invention.Multiple modification to these embodiment will be apparent concerning those skilled in the art, and defined General Principle can realize under the situation that does not break away from the spirit or scope of the present invention in other embodiments herein.Therefore, the present invention will can not be restricted to these embodiment shown in this article, but will meet the wideest scope consistent with principle disclosed herein and features of novelty.
Claims (3)
1. automatic hierarchy system of Kiwi berry, it is characterized in that, comprising: single file skate apparatus, single location conveying device, first motor, primary importance sensor, transparent transmission band, second motor, second place sensor, first camera, second camera, processor, controller and classification executive system; Wherein:
Described single file skate apparatus links to each other with described single location conveying device with storage device respectively, be used for to regulate the conveying capacity of the Kiwi berry that falls from storage device, makes it to become single-row shape and is delivered in the conveying device of described single location;
Described first motor links to each other with single location conveying device with described controller respectively, and the control signal that is used for sending according to described controller when classification drives described single location conveying device and carries Kiwi berry;
Described primary importance sensor is positioned at the conveying end of described single location conveying device, and link to each other with described controller, be used for when Kiwi berry arrives described single location delivery device tip, gathering the information that described Kiwi berry arrives, and the information that the described Kiwi berry that will gather arrives is sent to controller;
Described controller is controlled described first motor according to the information that receives and is shut down;
Described transparent conveyer belt links to each other with second motor with described single location conveying device respectively, is used for transmitting falling to the Kiwi berry of described transparent conveyer belt to image acquisition region from described single location conveying device;
Described second sensor is positioned at image acquisition region and links to each other with described controller, be used for when Kiwi berry arrives image acquisition region with described transparent conveyer belt, gathering described Kiwi berry information in place, and the described Kiwi berry information in place that will gather is sent to controller;
Described controller shuts down according to the second coupled motor of information control that receives, and the information that this Kiwi berry is in place sends to processor;
Described processor adopting triggers effective means, present frame Kiwi berry image in all Kiwi berry images of first camera in the image acquisition region and second camera collection is effective Kiwi berry image, and described effective Kiwi berry image is sent to image pre-processing module in the processor as handling object; Described first camera and second camera are gathered the described Kiwi berry image of two sides up and down respectively;
Described processor receives and handles described effective Kiwi berry image, according to result output classification results;
Described controller also links to each other with described processor, be used for to receive the classification results of described processor output, and controls described classification executive system according to described classification results Kiwi berry is dispensed to sorter frame;
Wherein, described processor comprises: human-computer interaction module, image pre-processing module, characteristic extracting module and feature Fusion Module; Wherein:
Described human-computer interaction module is by sending control command to controller, realize the controlling and monitoring of equipment, and coordinate image processing, IMAQ, controller three and work in order;
Described image pre-processing module is cut apart with the image cutting by image denoising, image the image of described first camera and second camera collection is carried out preliminary treatment; Described image denoising adopts small echo denoising mode;
Described characteristic extracting module is extracted size, shape and blemish feature from the pretreated image of process;
Described feature Fusion Module merges described size, shape and the blemish feature of extracting, and obtains classification results according to fusion results.
2. system according to claim 1 is characterized in that, described first camera links to each other with described processor by USB interface respectively with second camera.
3. system according to claim 1 is characterized in that, described controller is the STC89C52 single-chip microcomputer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210296845 CN102773217B (en) | 2012-08-20 | 2012-08-20 | Automatic grading system for kiwi fruits |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201210296845 CN102773217B (en) | 2012-08-20 | 2012-08-20 | Automatic grading system for kiwi fruits |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102773217A CN102773217A (en) | 2012-11-14 |
| CN102773217B true CN102773217B (en) | 2013-10-09 |
Family
ID=47118455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201210296845 Expired - Fee Related CN102773217B (en) | 2012-08-20 | 2012-08-20 | Automatic grading system for kiwi fruits |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102773217B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102921645A (en) * | 2012-11-19 | 2013-02-13 | 桂林电子科技大学 | Method and sorter for sorting momordica grosvenori based on image visual recognition |
| CN105032779B (en) * | 2015-07-15 | 2017-04-12 | 大连理工大学 | Accurate fishery organism detecting, counting and grading device |
| CN106290359B (en) * | 2016-07-22 | 2019-01-11 | 南京农业大学 | A kind of method of the lossless classification of apple crisp slices quality |
| CN106423905A (en) * | 2016-11-01 | 2017-02-22 | 无锡艾科瑞思产品设计与研究有限公司 | Kiwi detecting grading device |
| CN109261546A (en) * | 2017-07-18 | 2019-01-25 | 吕朝妮 | Kiwi berry automatic grading system |
| CN107694957A (en) * | 2017-09-29 | 2018-02-16 | 周永潮 | Automate feeding detection packing control method |
| CN108704867B (en) * | 2018-07-11 | 2023-12-12 | 四川农业大学 | Fruit grading plant |
| CN109816629B (en) * | 2018-12-20 | 2023-10-13 | 新绎健康科技有限公司 | Method and device for separating moss based on k-means clustering |
| CN111851028B (en) * | 2019-04-30 | 2023-05-12 | 青岛海尔洗衣机有限公司 | Clothes folding machine and control method thereof |
| CN110523657A (en) * | 2019-09-03 | 2019-12-03 | 广州能源检测研究院 | A kind of instrument appearance delection device and its working method based on 3D scanning method |
| CN115435689A (en) * | 2022-08-15 | 2022-12-06 | 北京旷视机器人技术有限公司 | Counterweight method, counterweight system, electronic equipment and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1666231A1 (en) * | 1989-04-06 | 1991-07-30 | Московский институт инженеров сельскохозяйственного производства им.В.П.Горячкина | Apparatus for sorting piecemeal objects |
| US5533628A (en) * | 1992-03-06 | 1996-07-09 | Agri Tech Incorporated | Method and apparatus for sorting objects by color including stable color transformation |
| CN101176872A (en) * | 2007-11-30 | 2008-05-14 | 华南理工大学 | Materials real-time detection and foreign body eliminating system based on machine vision |
| CN201176335Y (en) * | 2007-12-06 | 2009-01-07 | 朱壹 | Simple fruit high speed ordering conveying device for fruit vegetable sorting machine |
| CN202010662U (en) * | 2011-01-07 | 2011-10-19 | 昆明理工大学 | Real-time inspection and grading system device for fruit appearance quality |
-
2012
- 2012-08-20 CN CN 201210296845 patent/CN102773217B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1666231A1 (en) * | 1989-04-06 | 1991-07-30 | Московский институт инженеров сельскохозяйственного производства им.В.П.Горячкина | Apparatus for sorting piecemeal objects |
| US5533628A (en) * | 1992-03-06 | 1996-07-09 | Agri Tech Incorporated | Method and apparatus for sorting objects by color including stable color transformation |
| CN101176872A (en) * | 2007-11-30 | 2008-05-14 | 华南理工大学 | Materials real-time detection and foreign body eliminating system based on machine vision |
| CN201176335Y (en) * | 2007-12-06 | 2009-01-07 | 朱壹 | Simple fruit high speed ordering conveying device for fruit vegetable sorting machine |
| CN202010662U (en) * | 2011-01-07 | 2011-10-19 | 昆明理工大学 | Real-time inspection and grading system device for fruit appearance quality |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102773217A (en) | 2012-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102773217B (en) | Automatic grading system for kiwi fruits | |
| Qiang et al. | Identification of fruit and branch in natural scenes for citrus harvesting robot using machine vision and support vector machine | |
| Bresilla et al. | Single-shot convolution neural networks for real-time fruit detection within the tree | |
| CN206229716U (en) | A kind of apple exterior quality self-grading device based on computer vision technique | |
| Bechar et al. | Human‐robot collaboration for improved target recognition of agricultural robots | |
| Sun et al. | Image processing algorithms for infield single cotton boll counting and yield prediction | |
| CN113643280B (en) | Computer vision-based plate sorting system and method | |
| KR20200084940A (en) | CNN(Convolutional Neural Network) based pest and damage fruit classification device and method | |
| Koszela et al. | Computer image analysis in the quality in procedure for selected carrot varieties | |
| Chen et al. | A practical solution for ripe tomato recognition and localisation | |
| Yusuf et al. | Blob analysis for fruit recognition and detection | |
| CN115719451A (en) | Kiwi fruit actinidia arguta maturity detection method and system | |
| CN116363505A (en) | Target picking method based on picking robot vision system | |
| Raj et al. | Fruit Classification Comparison Based on CNN and YOLO | |
| CN103679144B (en) | Method for identifying fruits and vegetables in complex environment based on computer vision | |
| CN114462523A (en) | Prickly pear deburring and sorting integrated method based on image processing technology | |
| CN115147638A (en) | Machine vision-based cherry picking and classifying method and device | |
| Ignacio et al. | A YOLOv5-based deep learning model for in-situ detection and maturity grading of mango | |
| Khadabadi et al. | Disease detection in vegetables using image processing techniques: A review | |
| CN104484653A (en) | Bad corn kernel detecting method based on image recognition technology | |
| Gaikwad et al. | Fruit disease detection and classification | |
| Sidehabi et al. | The Development of Machine Vision System for Sorting Passion Fruit using Multi-Class Support Vector Machine. | |
| CN117456358A (en) | Method for detecting plant diseases and insect pests based on YOLOv5 neural network | |
| CN111369497B (en) | Walking type tree fruit continuous counting method and device | |
| Zahan et al. | A deep learning-based approach for mushroom diseases classification |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131009 Termination date: 20140820 |
|
| EXPY | Termination of patent right or utility model |