CN104899595A - Male and female silkworm chrysalis sorting and counting device based on SIFT (Scale Invariant Feature Transform) feature image - Google Patents
Male and female silkworm chrysalis sorting and counting device based on SIFT (Scale Invariant Feature Transform) feature image Download PDFInfo
- Publication number
- CN104899595A CN104899595A CN201510094468.6A CN201510094468A CN104899595A CN 104899595 A CN104899595 A CN 104899595A CN 201510094468 A CN201510094468 A CN 201510094468A CN 104899595 A CN104899595 A CN 104899595A
- Authority
- CN
- China
- Prior art keywords
- silkworm chrysalis
- male
- sorting
- female
- silkworm
- 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.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to a male and female silkworm chrysalis sorting and counting device based on an SIFT (Scale Invariant Feature Transform) feature image. The male and female silkworm chrysalis sorting and counting device adopts a video identifying and sorting device to favorably solve the problems that a hybrid rate is lowered and silkworm egg quality is lowered on an aspect of production due to the mixing of male and female silkworm chrysalises since errors are always caused by the manual identification and sorting of the silkworm chrysalises. The genders of the silkworm chrysalises can be automatically judged to drastically improve the labor efficiency of the identification and sorting of the silkworm chrysalises in China, a contradiction of labor force shortage in a seed production period is alleviated, meanwhile, male and female silkworm chrysalis identification errors can be reduced as much as possible, the silkworm egg hybrid rate is improved, and the silkworm egg quality in China can be drastically improved. The male and female silkworm chrysalis sorting and counting device can carry out automatic identification and sorting through a video image processing technology, can re-classify the silkworm chrysalises which can not be judged by a machine, solves a male and female silkworm chrysalis sorting problem through a mode of low cost and high efficiency and can save manpower and material resources so as to generate great economic benefit and social benefit.
Description
technical field:
the present invention relates to the silkworm chrysalis male and female sorting based on SIFT feature image and counting assembly, system image acquisition, analysis, processing module, and utilize the device of the design to achieve the Non-Destructive Testing of silkworm chrysalis sex.Result proves, detection method decreases the silkworm chrysalis wasting of resources and economic loss, can easy, intuitively, fast silkworm chrysalis is graded; Detection system measuring accuracy is high, for Fast nondestructive evaluation silkworm chrysalis sex has established theoretical and experiment basis.Proposing one utilizes digital image processing techniques based on area, and afterbody length breadth ratio and SIFT integration technology carry out resolving method.And can the number of accurate statistics silkworm chrysalis by device, result shows that precision is high, reliably easy, for Fast nondestructive evaluation silkworm chrysalis quality has established theoretical foundation.By gathering, analyzing tested silkworm chrysalis image, application image process software extrapolates fresh silkworm chrysalis weight automatically, finally carries out automatic sorting to silkworm chrysalis.
background technology:
sericulture, pupa silk industry as the typical chain type industry of China, always large with its industry chain length, market capacity, involve a wide range of knowledge, added value is high and famous.Silkworm chrysalis is the primary raw material of pupa silk industry.At present, China's silk industry product convergentization is obvious.Same Silkworm varieties, male silkworm chrysalis is compared with female silkworm chrysalis, and it is thin that the former has pupa silk size, the advantages such as cleanliness is excellent, and cohesive force is good, and raw silk standard is high, and fabric elasticity is good.Male and female silkworm chrysalis being utilized respectively, added value of product can be increased substantially when substantially not increasing Productive statistics, improve silk product quality, to the transformation that China realizes making the country prosperous from silk big country to silk, there is very important strategic importance.
silkworm chrysalis, also known as silkworm, be a kind of is foodstuff with mulberry leaf, has the insect that weaves silk of very high economic worth.Silkworm chrysalis is complete metamorphosis, and all one's life will through four diverse stages of development such as ovum, larva, pupa, adults.Ovum is that embryo occurs, grows and form stage of larva; Larva is pickuping food, the growth phase of repertory nutrition: pupa is the abnormal stage from larva to adult transition; The reproductive phase that adult is mate and oviposit, raise up seed.
silkworm chrysalis is the scrotiform protective seam of silkworm chrysalis formation in pupa time, includes pupal cell.Protective seam comprises the parts such as pupa clothing, pupa layer and pelettes, and pupa layer is the primary raw material of filature, and pupa clothing can be used as the raw material of silk flosssilk wadding and silk spinning.Silkworm chrysalis has the difformity such as ellipse, elliptical beam kidney-shaped, spherical or spindle, and the shape difference of female, male silkworm chrysalis is less, is difficult to rely on silkworm chrysalis shape facility to judge silkworm chrysalis sex.
silkworm chrysalis larva shortly can become silkworm chrysalis after being placed on small straw bundles to spin cocoons and tying pupa.The similar spindle of the silkworm chrysalis bodily form, separately, chest, the individual section of abdomen three.Head length has compound eye and feeler; Chest is long pereiopoda and wing; Belly merogenesis.Female pupa belly is large and end is round, and there is an ordinate octoon outside of belly center; Male pupa belly is little and end is sharp, and there are a brown point in the 9th uromere outside of belly central authorities.Female, male silkworm chrysalis shape difference is comparatively large, can judge silkworm chrysalis sex by silkworm chrysalis morphological feature difference.
summary of the invention:
the object of this invention is to provide the silkworm chrysalis male and female sorting based on SIFT feature image and counting assembly.
the Sexual discriminating of 1 silkworm chrysalis
image semantic classification comprises the processes such as image enhaucament, image denoising, Iamge Segmentation, edge extracting, is the low level processing procedure of image recognition and understanding.In the research of silkworm chrysalis Sex Discrimination, the Accuracy and high efficiency of net result depends on the quality of Image semantic classification to a great extent.
1) based on the Otsu multi thresholds method of histogram search
otsu threshold method belongs to the searching method of a kind of global optimum threshold value, when the variance difference of two classes in image is too large, the mean distance of segmentation threshold and surging class can be made too near and cause classifying unsuccessfully, claiming this image to be that Otsu is inseparable in this article.Silkworm chrysalis region must meet the superimposed characteristics (may be presented as unimodal, multimodal or impulse form) of the normal distribution with obvious crest, therefore, in grey level histogram, must there is a certain or certain several peak gray and belong to silkworm chrysalis region.Based on above-mentioned analysis, will start with from grey level histogram, solve the inseparable problem of Otsu of gray level image.
definition 1: in specific region, specific tonal range, be the number of the pixel at edge and the ratio of region populations number of pixels by particular edge operator identification, be called marginal density, represent with d; The average of all marginal point gradient magnitudes is then called edge strength, represents with A.In this article, marginal density and edge strength usually together be used to weigh the texture features of specific region.
theorem 1: in grey level histogram, at least exists a bit
, must exist a certain field (
), make the arbitrfary point in its neighborhood
meet
set up.
prove: be the gray level image of L for gray level, if be a certain neighborhood of 0 in gray scale
interior arbitrfary point
all satisfied
set up, problem must be demonstrate,proved; Otherwise, in field
in have at least a bit
meet
set up;
for
if, at neighborhood
arbitrfary point meet
set up, problem must be demonstrate,proved; Otherwise, in field
in have at least 1 n meet
set up;
by that analogy, because the span of gray scale is limited, must and also at least exist 1 i, to its a certain neighborhood (
) in arbitrfary point j meet
set up.Card is finished.
theorem 1 illustrates peak value must be had to exist in grey level histogram.And the gray scale of peak value is very likely distributed near the average in silkworm chrysalis region.Assuming that n the peak value detected in t is expressed as ,
wherein for peak value gray scale, for peak value height average in a certain neighborhood, with be respectively the marginal density in a certain neighborhood of peak value and edge strength, then peak value the probability belonging to silkworm chrysalis region is expressed as:
（1）
wherein
,
with
be respectively the peak-peak gray scale of jth quasi-mode in reference zone in image, marginal density and edge strength.Consider the adverse effect of high light, shade and silkworm chrysalis defect, at most reference zone is divided into 2 classes.Reference zone selects the middle and lower part at image usually, and shape is triangle or trapezoidal, reference zone herein by some homalographic rectangular blocks form trapezoidal, each rectangular block is classified according to peak-peak gray scale, marginal density and edge strength 3 kinds of features.The observation model of formula (1) right-hand member is used for calculating the degree of closeness of a certain class in peak value i in present image histogram and reference zone.Equation right-hand member motion model, then the peak value similarity of consecutive frame image is associated, combine the texture features (marginal density and edge strength) of peak gray, peak height and peak ranges, the accuracy that the various features in silkworm chrysalis region are roughly estimated can be improved.Wherein
for normalized parameter, make the value of probability between 0 and 1.
2) peak value cluster and multi-threshold segmentation
because silkworm chrysalis region may be presented as that multimodal distributes, right
carry out descending sort.The peak value of maximum probability is chosen for cluster centre, utilizes the similarity of marginal density and edge strength between peak value to carry out peak value cluster.Assuming that cluster centre is peak value i, then peak value j and peak value i can gather for of a sort decision criteria is: the similarity of peak value i and peak value j marginal density and edge strength is greater than specific threshold, and all peak values between i and j are all similar with peak value i.Otsu formula (2) between any two points is out of shape to obtain, obviously, in formula to formula (1)
,
,
,
with
in
interval all to adjust accordingly.
（2）
above-mentioned algorithm makes by means of iterative operation
progressively away from the region of high peaks, when alternative manner lost efficacy, then relied on the result of histogram analysis.
3) SIFT feature coupling
sIFT feature coupling compares two profiler, if the distance of two profiler is less than given threshold value, just claims these two features mutually to mate, otherwise claims these two features not mate.The method of characteristic matching has a lot, but most convenient, the most frequently used method are nearest neighbor algorithms, also will adopt the coupling realized in this way between different images scene characteristic herein.Provide the mathematical description of most neighbor point below: suppose that the field of definition of a K dimension space and codomain are expressed as R and D, S is the set of all sampled points in K dimension space R × D, and d is the vector of impact point, then the most neighbor point of d
the condition of formula (5-8) must be met:
(5-8)
in formula (5-8), di is the i-th dimension component of vectorial d.One directly, simple algorithm be to calculate successively in S the distance a little and between impact point d, thus find the point the most adjacent with d.The time complexity of this algorithm is O (N), and wherein N is the number of unique point in S.But unique point needs to carry out a large amount of couplings in silkworm chrysalis detects, in order to improve the efficiency of coupling, use the most neighbor point searching algorithm based on K-D tree (K-D tree) [162-163] herein, time complexity can be reduced to O (log2N) by it.
distance between the SIFT feature point Euclidean distance of its describer represents.SIFT feature coupling realizes by mating the describer of unique point.Nearest neighbor algorithm based on K-D tree all SIFT feature observed is set up K-D tree, and to each SIFT feature kp that Current observation arrives, and finds most neighbor point kp1 and secondary neighbor point kp2 with most Proximal Point Algorithm set based on K-D.Experiment shows, if |
| less and |
| larger, so the quality of kp and kp1 coupling is higher.If therefore met:
(5-9)
then think that kp and kp1 mates, wherein
for the constant of boundary between 0 and 1.In this article
=0.75, this eliminates the feature point pairs that some matching errors are larger.
it is the real time problems of feature extraction and matching that SIFT feature is applied to the biggest problem that silkworm chrysalis detects.Based on K-D tree nearest neighbor search algorithm the time complexity of characteristic matching from
reduce to
, wherein N is the number of element in sample space.Time due to feature extraction mainly consumes at gray level image in the convolution of different proportion gaussian kernel, although the width reducing image resolution ratio or reduction convolution mask can reduce computing, to a certain extent to reduce correct matching rate for cost.
this research institute CCD camera imaging technical limit spacing image, first gaussian filtering is carried out, image after denoising, according to gradation of image distribution characteristics, the silkworm chrysalis image collected is carried out the segmentation of OTSU method again, carry out data fusion by the major axis and minor axis value ratio and SIFT feature extraction etc. of splitting rear silkworm chrysalis region afterbody again, judged the sex of silkworm chrysalis by average classification.The weight data of each silkworm chrysalis also can according to by silkworm chrysalis segmentation, and the mode of weighting, calculate the weight of individual silkworm chrysalis.
2 silkworm chrysalis male and female sorting equipments
by 1) silkworm chrysalis feed arrangement, silkworm chrysalis is distinguished one by one, fall 2) wind-controlling device used by travelling belt, silkworm chrysalis is delivered to 5) baffle plate place and by 4) infrared probe detects whether has silkworm chrysalis, if there is triggering 6) video camera up and down in video identification equipment identification device carries out collection image, tempering clear glass material is adopted in 6, so, all can collect silkworm chrysalis image clearly up and down, after carrying out image processing and analyzing, if male silkworm chrysalis, then mechanical arm 3) push silkworm chrysalis to 8) male silkworm chrysalis gatherer, otherwise, by travelling belt 2) deliver to 7) in female silkworm chrysalis gatherer, if cannot differentiate, this controls 9 by industrial computer) plectrum to be allocated in 7 right side, otherwise be allocated to left side, sort complete.
wherein, the number of silkworm chrysalis can be counted by mechanical arm and infrared ray.Pass through, video identification device can differentiate the sex of silkworm chrysalis.Can be sorted it by travelling belt and mechanical arm.Filled with by DSP module embedded in control panel and write resolving method.
beneficial effect:
(1) information can be reviewed: due to can the information such as Online statistics number, time, the place of production, weight to the silkworm chrysalis of each batch, convenient management.
(2) utilization factor is high: electronic equipment replaces manual work, greatly increases work efficiency, and saves cost of labor.
(3) security is high: smart machine operation, and information can store can trace to the source, and also can be applied in shopping online experience, the destructive power of silkworm chrysalis is little.
accompanying drawing illustrates:
accompanying drawing 1 is silkworm chrysalis male and female sorting equipments
1) silkworm chrysalis feed arrangement 2) conveyer 3) mechanical pushing hands 4) infrared probe 5) baffle plate 6) video identification equipment identification device, for transparent material, video camera 7 is housed up and down) female silkworm chrysalis gatherer 8) male silkworm chrysalis gatherer 9) sorting plectrum 10) microcontroller 11) display
fig. 2 silkworm chrysalis image processing flow figure
embodiment:
1. based on the sorting of silkworm chrysalis male and female and the counting assembly of SIFT feature image, its composition comprises: 1) silkworm chrysalis feed arrangement 2) conveyer 3) mechanical pushing hands 4) infrared probe 5) baffle plate 6) video identification equipment identification device, for transparent material, video camera 7 is housed up and down) female silkworm chrysalis gatherer 8) male silkworm chrysalis gatherer 9) sorting plectrum 10) microcontroller 11) critical piece such as display composition.
2. the sorting of the silkworm chrysalis male and female based on SIFT feature image according to claim 1 and counting assembly, is characterized in that: described microcontroller is based on 16 bit CPU process chip expansion board.
3. the sorting of the silkworm chrysalis male and female based on SIFT feature image according to claim 1 and 2 and counting assembly, it is characterized in that by 1) silkworm chrysalis feed arrangement, silkworm chrysalis is distinguished one by one, fall 2) wind-controlling device used by travelling belt, silkworm chrysalis is delivered to 5) baffle plate place and by 4) infrared probe detects whether has silkworm chrysalis, if there is triggering 6) video camera up and down in video identification equipment identification device carries out collection image, tempering clear glass material is adopted in 6, so, all can collect silkworm chrysalis image clearly up and down, after carrying out image processing and analyzing, if male silkworm chrysalis, then mechanical arm 3) push silkworm chrysalis to 8) male silkworm chrysalis gatherer, otherwise, by travelling belt 2) deliver to 7) in female silkworm chrysalis gatherer, if cannot differentiate, this is by microprocessor controls 9) plectrum to be allocated in 7 right side, otherwise be allocated to left side, sort complete.
4. the sorting of the silkworm chrysalis male and female based on SIFT feature image according to claim 1 or 2 or 3 and counting assembly, it identifies to differentiate and controls trigger pip by based on area, and afterbody length breadth ratio and SIFT integration technology trigger.
the foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (4)
1. based on the sorting of silkworm chrysalis male and female and the counting assembly of SIFT feature image, its composition comprises: 1) silkworm chrysalis feed arrangement 2) conveyer 3) mechanical pushing hands 4) infrared probe 5) baffle plate 6) video identification equipment identification device, for transparent material, video camera 7 is housed up and down) female silkworm chrysalis gatherer 8) male silkworm chrysalis gatherer 9) sorting plectrum 10) microcontroller 11) critical piece such as display composition.
2. the sorting of the silkworm chrysalis male and female based on SIFT feature image according to claim 1 and counting assembly, is characterized in that: described microcontroller is based on 16 bit CPU process chip expansion board.
3. the sorting of the silkworm chrysalis male and female based on SIFT feature image according to claim 1 and 2 and counting assembly, it is characterized in that by 1) silkworm chrysalis feed arrangement, silkworm chrysalis is distinguished one by one, fall 2) wind-controlling device used by travelling belt, silkworm chrysalis is delivered to 5) baffle plate place and by 4) infrared probe detects whether has silkworm chrysalis, if there is triggering 6) video camera up and down in video identification equipment identification device carries out collection image, tempering clear glass material is adopted in 6, so, all can collect silkworm chrysalis image clearly up and down, after carrying out image processing and analyzing, if male silkworm chrysalis, then mechanical arm 3) push silkworm chrysalis to 8) male silkworm chrysalis gatherer, otherwise, by travelling belt 2) deliver to 7) in female silkworm chrysalis gatherer, if cannot differentiate, this is by microprocessor controls 9) plectrum to be allocated in 7 right side, otherwise be allocated to left side, sort complete.
4. the sorting of the silkworm chrysalis male and female based on SIFT feature image according to claim 1 or 2 or 3 and counting assembly, it identifies to differentiate and controls trigger pip by based on area, afterbody length breadth ratio and SIFT integration technology are differentiated, and counting triggering has infrared induction signal to complete.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510094468.6A CN104899595A (en) | 2015-03-04 | 2015-03-04 | Male and female silkworm chrysalis sorting and counting device based on SIFT (Scale Invariant Feature Transform) feature image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510094468.6A CN104899595A (en) | 2015-03-04 | 2015-03-04 | Male and female silkworm chrysalis sorting and counting device based on SIFT (Scale Invariant Feature Transform) feature image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104899595A true CN104899595A (en) | 2015-09-09 |
Family
ID=54032249
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510094468.6A Pending CN104899595A (en) | 2015-03-04 | 2015-03-04 | Male and female silkworm chrysalis sorting and counting device based on SIFT (Scale Invariant Feature Transform) feature image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104899595A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105666485A (en) * | 2016-03-28 | 2016-06-15 | 桂林电子科技大学 | Automatic identifying and positioning chess placing robot based on image processing |
| CN106035251A (en) * | 2016-06-30 | 2016-10-26 | 霍山县天衡桑蚕制种有限公司 | Silkworm pupae sex identification method |
| CN106198448A (en) * | 2016-07-17 | 2016-12-07 | 北京化工大学 | A kind of automatic high speed lossless sorting live body male and female Pupa bombycis or the technique of live body male and female Bombyx bombycis |
| CN108522437A (en) * | 2017-05-08 | 2018-09-14 | 杭州蚕鱼科技有限公司 | Full-automatic male and female silk cocoon sorting unit |
| CN109499889A (en) * | 2018-11-20 | 2019-03-22 | 漯河医学高等专科学校 | A kind of campus library book automatic identification sorting device |
| CN112085142A (en) * | 2020-08-29 | 2020-12-15 | 中国石油大学(华东) | Rapid counting device and counting method for large-batch fish fries |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101403142A (en) * | 2008-11-04 | 2009-04-08 | 江苏苏豪国际集团股份有限公司 | Automatic male and female sorting apparatus for cocoons judged by fluorescence color |
| CN102095481A (en) * | 2010-12-02 | 2011-06-15 | 江苏大学 | Device and method for automatically discriminating cocoon sexuality |
| CN202247028U (en) * | 2011-07-26 | 2012-05-30 | 宏宝集团金寨丝绸有限公司 | Rotary cocoon collecting device |
| CN202925160U (en) * | 2012-11-18 | 2013-05-08 | 沾益县中兴茧丝绸实业有限公司 | Silkworm cocoon sorting device |
-
2015
- 2015-03-04 CN CN201510094468.6A patent/CN104899595A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101403142A (en) * | 2008-11-04 | 2009-04-08 | 江苏苏豪国际集团股份有限公司 | Automatic male and female sorting apparatus for cocoons judged by fluorescence color |
| CN102095481A (en) * | 2010-12-02 | 2011-06-15 | 江苏大学 | Device and method for automatically discriminating cocoon sexuality |
| CN202247028U (en) * | 2011-07-26 | 2012-05-30 | 宏宝集团金寨丝绸有限公司 | Rotary cocoon collecting device |
| CN202925160U (en) * | 2012-11-18 | 2013-05-08 | 沾益县中兴茧丝绸实业有限公司 | Silkworm cocoon sorting device |
Non-Patent Citations (2)
| Title |
|---|
| 蔡健荣等: "X射线成像技术判别蚕茧性别", 《农业工程学报》 * |
| 龚攀: "基于计算机视觉的蚕蛹性别自动识别系统研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105666485A (en) * | 2016-03-28 | 2016-06-15 | 桂林电子科技大学 | Automatic identifying and positioning chess placing robot based on image processing |
| CN105666485B (en) * | 2016-03-28 | 2018-05-11 | 桂林电子科技大学 | A kind of automatic identification based on image procossing and positioning Bai Qi robots |
| CN106035251A (en) * | 2016-06-30 | 2016-10-26 | 霍山县天衡桑蚕制种有限公司 | Silkworm pupae sex identification method |
| CN106198448A (en) * | 2016-07-17 | 2016-12-07 | 北京化工大学 | A kind of automatic high speed lossless sorting live body male and female Pupa bombycis or the technique of live body male and female Bombyx bombycis |
| CN108522437A (en) * | 2017-05-08 | 2018-09-14 | 杭州蚕鱼科技有限公司 | Full-automatic male and female silk cocoon sorting unit |
| CN108522437B (en) * | 2017-05-08 | 2020-10-02 | 杭州蚕鱼科技有限公司 | Full-automatic male and female silkworm cocoon sorting unit |
| CN109499889A (en) * | 2018-11-20 | 2019-03-22 | 漯河医学高等专科学校 | A kind of campus library book automatic identification sorting device |
| CN112085142A (en) * | 2020-08-29 | 2020-12-15 | 中国石油大学(华东) | Rapid counting device and counting method for large-batch fish fries |
| CN112085142B (en) * | 2020-08-29 | 2023-09-19 | 中国石油大学(华东) | Rapid counting device and counting method for large-batch fries |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104899595A (en) | Male and female silkworm chrysalis sorting and counting device based on SIFT (Scale Invariant Feature Transform) feature image | |
| Aravind et al. | Disease classification in maize crop using bag of features and multiclass support vector machine | |
| CN107016405B (en) | A kind of pest image classification method based on classification prediction convolutional neural networks | |
| CN102842032B (en) | Method for recognizing pornography images on mobile Internet based on multi-mode combinational strategy | |
| Qing et al. | Automated detection and identification of white-backed planthoppers in paddy fields using image processing | |
| CN114723704B (en) | Textile quality evaluation method based on image processing | |
| CN111861103A (en) | Fresh tea leaf classification method based on multiple features and multiple classifiers | |
| CN109886277B (en) | Contour analysis-based fresh tea leaf identification method | |
| CN107341688A (en) | The acquisition method and system of a kind of customer experience | |
| CN102184405A (en) | Image acquisition-analysis method | |
| CN105513066B (en) | It is a kind of that the generic object detection method merged with super-pixel is chosen based on seed point | |
| Zhang et al. | Real-time monitoring of optimum timing for harvesting fresh tea leaves based on machine vision | |
| CN111940339B (en) | Red date letter sorting system based on artificial intelligence | |
| CN108154195A (en) | Tealeaves recognition methods and the tealeaves sorting equipment using this method | |
| CN112893159B (en) | Coal gangue sorting method based on image recognition | |
| CN106651882A (en) | Method and device for identifying and detecting cubilose impurities based on machine vision | |
| CN109886932A (en) | Gear ring of wheel speed sensor detection method of surface flaw based on SVM | |
| Zhou et al. | Research on cucumber downy mildew detection system based on SVM classification algorithm | |
| Deulkar et al. | An automated tomato quality grading using clustering based support vector machine | |
| CN106326914B (en) | A kind of more classification methods of pearl based on SVM | |
| Yang et al. | Automatic greenhouse pest recognition based on multiple color space features | |
| CN107403162A (en) | The data acquisition of vehicle notification number and the system and method for classification | |
| Bai et al. | Recognition of bovine milk somatic cells based on multi-feature extraction and a GBDT-AdaBoost fusion model | |
| CN114985283B (en) | Tea leaf identification method based on multi-feature image identification technology | |
| Alyas et al. | Detection of plant diseases using image processing with machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150909 |
|
| WD01 | Invention patent application deemed withdrawn after publication |