CN107680098A - A kind of recognition methods of sugarcane sugarcane section feature - Google Patents

A kind of recognition methods of sugarcane sugarcane section feature Download PDF

Info

Publication number
CN107680098A
CN107680098A CN201711061414.5A CN201711061414A CN107680098A CN 107680098 A CN107680098 A CN 107680098A CN 201711061414 A CN201711061414 A CN 201711061414A CN 107680098 A CN107680098 A CN 107680098A
Authority
CN
China
Prior art keywords
sugarcane
image
threshold value
section
carried out
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
Application number
CN201711061414.5A
Other languages
Chinese (zh)
Inventor
李尚平
鲁碧辉
周永权
文春明
陈增雄
李凯华
张可
王梦萍
莫德庆
王旭艳
袁泓磊
李尚辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangxi University for Nationalities
Original Assignee
Guangxi University for Nationalities
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangxi University for Nationalities filed Critical Guangxi University for Nationalities
Priority to CN201711061414.5A priority Critical patent/CN107680098A/en
Publication of CN107680098A publication Critical patent/CN107680098A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Molecular Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Image Processing (AREA)

Abstract

The present invention provides a kind of recognition methods of sugarcane sugarcane section feature, belongs to agricultural equipment intelligent identification Method field, this method comprises the following steps：Collection to identifying image；The series of preprocessing such as denoising are carried out to the image of collection；Cluster analysis is carried out to image；Carry out gray value average computation；The determination of threshold value is carried out using crow swarm intelligence algorithm；Segmentation figure picture；Calculate and extract part interested (sugarcane section)；The detection of accuracy rate is identified；Result is stored in database.This method takes crow colony intelligence optimized algorithm to improve the intelligent decision to image threshold, and accurate segmentation figure picture extracts the position of sugarcane section, and establishes database and experimental result data is stored in into database analysis, to improve the accuracy rate of identification.Rapidly and accurately sugarcane sugarcane section position is identified the present invention seeks to realize, the automatic of kind of machine is cut for follow-up intelligence and cuts kind of an operation technical support is provided.

Description

A kind of recognition methods of sugarcane sugarcane section feature
Technical field
The present invention relates to agricultural equipment intelligent identification Method field, is specifically related to a kind of sugarcane sugarcane section feature Recognition methods.
Background technology
Guangxi is the major production areas of national cane suger, accounts for 60%, but due to sugarcane field many places knob, mechanization Degree is than relatively low, and the plantation of sugarcane at present is main, and also manually plantation, harvest operation, artificial cost are too high.In recent years, due to Being declined by sugared valency is influenceed, and sugared valency drops to 4500 yuan/ton by 7080 yuan/ton, and sugarcane procurement price is even more by 500 yuan/ton of declines To 400 yuan/ton, until last year has just risen again so that the sugarcane yield in Guangxi has declined, and the prices of sugarcane purchase The income of peasant is directly affects, its reason is that mechanization degree is too low, causes cost too high.Last year China mechanization association A grand sugarcane whole process production mechanization seminar has been held in Liuzhou, put on display the current country also continually develop it is a variety of sweet Sugarcane planting machine, sugarcane harvesting ploughing machine, Irrigation Machine etc., further promote the paces of Guangxi mechanization production plantation.But, In recent years mechanization is also incorporated into field by Guangxi various regions in succession, but at present still based on whole bar in real time cut-out, between needing 4 people not The operation of disconnected ground, not only labor intensity is big, and due to being measured length cutting sugarcane kind, easily hinders sugarcane section, and can not judge sugarcane section Quality, with more (800-1000kg/ mus) is planted, and bud ratio can not be ensured；It is badly in need of exploitation sugarcane pre-cut kind technology at present, its is prior Cutting screening, processing are carried out to sugarcane kind, it is few (250-300kg/ mus) with kind, and high bud ratio can be ensured, but for machine In terms of the identification for the pre-cut kind that tool grows cane, also in the presence of certain technical barrier.The systematic research is developed so that a lot The sugarcane section identification problem that kind is cut in intellectuality is resolved, and the plantation of mechanization pre-cut kind is become a reality, therefore realize machine intelligence The feature recognition of the sugarcane section of sugarcane sugarcane kind can be identified, is furtherd investigate, is beneficial in terms of color model and Intelligent Recognition The key technical problem of sugarcane pest and disease damage sugarcane section feature recognition is cracked, can be as the reason of research cane planting intelligence pre-cut kind equipment By foundation.In the future it is also possible to developing toward the direction more modernized.Reduce the amount of labour and mean that the cost for reducing plantation, into Originally lack economic benefit just to improve naturally, be advantageous to promote pre-cut kind technology and the mechanized cultivation application of sugarcane of sugarcane.
Domestic present Research:
There are many breakthrough progress in the research of sugarcane planting machine.Sweet 75-1 types planting machine can carry out cane planting Entire mechanization, it is by China National Light Industrial Products Department's Sugar Industry Research Institute.The type sugarcane joint of Guangdong -2 that Guangdong Province's agricultural machinery is developed The work such as planting machine can be ditched, cut kind, row are planted, watered, banketing.Also there is the research to sugarcane planting machine in Guangxi, extensively The rich 2CZ-1 sugarcane combined planters of celebrating of western Zhuang autonomous region agricultural machinery Research Institute, planting machine, which combines, to be started, is driven, sending Cut kind, the function such as powder-spreading and sterilizing, row are planted, applied fertilizer, banketing.
Also there is the much research on machine vision in terms of other of agricultural.Lu Shang equalitys are with fuzzy decision to sugarcane Kind identification is studied, and also has the decision-making set of this kind of Algorithm constitution of edge fitting for the component HSV and RGB of color, With the outstanding algorithm of fuzzy clustering, then the component combination of threshold condition is met into a set, by fuzzy subset's element most As stipes section class, this method is very high for the discrimination of stipes, but is still in theoretical research stage for more classification.Zhao Enter the researchers such as brightness to be identified for sugarcane pest and disease damage by sugarcane vein picture, have developed suitable mountain region, knob, The less sugarcane planting machine of land area.The sugarcane Lal that the personnel such as Guo Lin set up with support vector regression method estimates mould Type, and used relatively new type domestic deep space satellite data environment star CCD pictures and quasi synchronous ground observation data The regression model of foundation, in order to propose a kind of novel think of using the method inverting Sugarcane Leaves area index of remote control sensing Road.It is distant that the personnel such as Kuang Zhaomin using vegetation state indices (VCD) and temperature condition index (TCD) establish drought index (DID) The model of induction monitoring is controlled, moreover, is also merged the ETM satellite datas with high spatial resolution so that distant The detection method of control sensing can apply to sugarcane arid this respect..The personnel such as Tan Zongkun utilize and calculate Guangxi province multidate MODIS-NDVI values, the change curve of the sample training area multidate NDVI values of sugarcane is selected with reference to GPS, by sensing to remote control Compared with image sugarcane contiguous plant section, scattered growing area with GPS position investigation on the spot, it was demonstrated that the cane planting of Guangxi province Information remote control induction monitoring is consistent with actual state.Liu Qingting etc. cuts cane stalk section using High-speed Photography Analysis blade and destroyed The journey researchs to sugarcane image in China at present have focused largely on sugarcane yield assessment and the disease region detection of macroscopic view.To sweet The research of machine vision in sugarcane mechanical development is also fewer, and the present invention can make up this vacancy.
External many research workers already apply to identification technology the every aspect of agricultural.In terms of cane planting, Except China, there is the country such as Australia, Brazil, Cuba kind to be implanted with more sugarcane in the world, these countries are in sugarcane kind Mechanization degree in plant is all higher.External common sugarcane planting machine mainly has three kinds, whole stalk formula sugarcane planting machine, in real time Kind of formula sugarcane planting machine and pre-cut kind formula sugarcane planting machine are cut, but image recognition rarely has the utilization in terms of cane planting machinery.
In terms of image recognition, at one in the research of peach, in order that peach automatic classification, Miller B K etc. The experiment that the method that people is split using image is carried out, summarize a kind of method for the area for calculating peach surface damage, the technology Core be exactly near-infrared image analysis.The research application on apple surface scar that Rehkugler et al. is done arrives The method of gray scale.Sarkar N and Wolfe R R et al. are in a size on studying tomato and shape and damage In work, the methods of they have used Digital Image Processing, the surface defect of fruit is carried out with a very special method Identification, that is, the method for shade of gray curve.Paulsen et al. one for gradation of image value histogram in different background Under distribution, it is proposed that contrast has shown that contrast is higher with the relation between corresponding two peak distances on gray-scale map, away from From bigger conclusion.Wan DeVooren et al. are at one in the research of mushroom, having used machine vision technique to its shape The factors such as shape are identified, it is proposed that on obtaining the algorithm of shape facility from green crop so that discrimination reaches 69%. At one in the research of egg, the crackle on egg surface is identified with machine vision technique, rate of accuracy reached arrives Goodrum et al. 94%.
The content of the invention
The purpose of the present invention proposes a kind of recognition methods of sugarcane sugarcane section feature, the equipment available for sugarcane pre-cut kind Intelligent automatic identification, automation cutting is carried out to sugarcane.
The present invention solves the above problems by the following technical programs：
A kind of recognition methods of sugarcane sugarcane section feature, comprises the following steps：
Step 1：The sugarcane for choosing different cultivars carries out IMAQ, obtains gathering image；
Step 2：Denoising is carried out to collection image and gray processing handles to obtain pretreatment image；
Step 3：European cluster analysis is carried out to pretreatment image to obtain analyzing image；
Step 4：Gray value average computation is carried out to analysis image；
Step 5：Computing is carried out to the image after calculating gray value using crow colony intelligence optimized algorithm and determines sugarcane sugarcane The characteristic threshold value of section；
Step 6：Dividing processing is carried out to image according to the characteristic threshold value of the sugarcane sugarcane section in step 5；
Step 7：Sugarcane sugarcane section information extraction is carried out to the image after segmentation；
Step 8：The sugarcane sugarcane section information of extraction is accurately detected, if accurately, into next step；It is if inaccurate Really, then the 5th step is returned；
Step 9：It is R reference threshold in threshold value and databaseoCarry out computing, Ri-Ro>=M, M obtain for lot of experimental data The error gone out, RiFor threshold value, if meeting that condition is put in storage, otherwise give up.
After being come out by threshold method to each sugarcane sugarcane section Automatic feature recognition, pass through the pixel of sampling Point obtains the size positions of sugarcane sugarcane section and the specific length respectively saved, and input database compared with the length dimension demarcated, And it is supplied to kind of machine of cutting to carry out cutting kind of an operation.
It is the shape to variety classes sugarcane to the process of sugarcane IMAQ preferably in step 1 in such scheme IMAQ is carried out with color.
In such scheme, the detailed process that the characteristic threshold value of sugarcane sugarcane section is determined preferably in step 5 is：By image its In one or N number of pixel as a particle, computing is done using crow colony intelligence optimized algorithm and finds a gray value Characteristic threshold value of the scope as the sugarcane sugarcane section of reference, and judges whether this threshold value is suitable, if appropriate then threshold value, enters Enter in next step, if improper return to step 4 recalculates searching threshold value.
In such scheme, the detailed process that image is split preferably in step 6 is：Purpose sign gray value is set to 0 i.e. For black, it is white that other positions are uniformly set to gray value 255, and black and white segmentation is carried out to image according to threshold values.
In such scheme, the process of sugarcane sugarcane section information extraction is preferably in step 7：Picture black part is carried out Information extraction.
Advantages of the present invention is with effect：
The present invention saves automatic identification by carrying out sugarcane to sugarcane seed, so that sugarcane cutting equipment can be automatic That changes is accurately cut to sugarcane seed, realizes the sugarcane seed cutting of automation, and by adding crow colony intelligence Optimized algorithm (CSA) carries out fast and accurately intelligent decision to the threshold value of sugarcane sugarcane section feature.The determination of threshold value can be more preferable Segmentation figure picture, and the accuracy rate of characteristics of image identification is improved, automatically extracting for intelligentized sugarcane sugarcane section feature is carried out to realize, The automatic of kind of machine is cut for follow-up intelligence to cut kind of an operation and lay the foundation.
Brief description of the drawings
Fig. 1 is flow chart of the present invention.
Embodiment
The invention will be further described with reference to embodiments.
The principle of crow colony intelligence optimized algorithm (CSA) is as follows：
Crow is lived in the form of population.
Crow remembers that oneself hides the position of food.
Crow follows other side to do stealer.
Crow protects the Tibetan food position of oneself with certain probability.
It suppose there is in D dimension spaces, the quantity (size of population) of crow is N and crow the t generations in search space Position is xi,iter(i=1,2,3 ... N；Iter=1,2 ..., iterfmax), iterfmaxFor maximum iteration.Every crow There is a memory in place of hiding food to it, in the i-th ter generations, the position of crow i Tibetan food is mi,iter.In the i-th ter generations, mi,iterIndicate i-th crow to iter instead of it is preceding hide food best position.In fact, each best Tibetan of crow The position of food is remembered.Crow is moved in D dimension spaces, finds more preferable food source (covert).
Assuming that in the i-th ter generations, crow desires access to its Tibetan food, and crow i determines to follow crow j to go to find crow j Tibetan In place of food.In this case, there are two kinds of shape possible states：
State 1：Crow J does not know that crow i follows it.Therefore, crow i can be close in place of crow j Tibetan food.This In the case of, crow i new position is as follows：
xi,iter+1=xi,iter+ri×fli,iter×(mi,iter-xi,iter)
(1)
riEqually distributed random number, fl between 0 and 1i.iterRefer to flying distances of the crow i in the i-th ter generations.
State 2：Crow J knows that crow i follows it.Therefore, it is stolen to prevent from hiding food position, crow j is by crow i Lead for another position.
Generally speaking, state 1 and 2 can represent as follows：
riEqually distributed random number, AP between 0 and 1i,iterRepresent consciousness probability of the crow j in the i-th ter generations.
Meta-heuristic algorithm should provide good balance between global search and Local Search.In CSA, the overall situation is searched Rope and Local Search are mainly by the control of consciousness probability (AP) parameter.Tended to by reduction consciousness probable value AP, CSA in part Region scans for.Therefore, Local Search can be increased using small AP values.On the other hand, by improving perceived probability value AP, search The probability of the solution of Suo Dangqian optimal locations reduces, and CSA tends to explore search space with global yardstick (randomization).Cause This, can increase diversification using larger AP values.During threshold value is optimized, each crow represents a kind of threshold optimization Scheme, optimal threshold value allocative decision is determined according to position iterative formula above, until reaching maximum iterations.According to After crow group's algorithm determines optimal threshold value allocative decision, image segmentation is carried out.
Identification process：
IMAQ:Choose the sugarcane progress IMAQ of a large amount of different cultivars sugarcane shapes, color early stage, to reaching, And the picture of clear sugarcane is taken using professional camera.
Image preprocessing：Denoising is carried out for the picture obtained in the 1st step, so as to remove some easy shadows in background The factor of extraction feature is rung, and handled good picture is subjected to gray processing processing.
Cluster analysis：The image that second step is obtained carries out European cluster analysis, so as to by one on sugarcane after gray scale The higher point of a little gray values (not sugarcane section) is gathered in purpose feature, improves recognition accuracy.
Gray value mean value calculation：By whole picture with matlab calculating average gray value, for the judgement of the 5th step threshold value Prepare.
The calculating and judgement of threshold value：Using crow colony intelligence optimized algorithm threshold value, we are by some in picture Either certain multiple pixel does computing using crow colony intelligence optimized algorithm and finds an intensity value ranges as a particle As the characteristic threshold value of the sugarcane sugarcane section of reference, and judge whether this threshold value is suitable.Then threshold value if appropriate, and carry out 6th step, if improper recalculate searching threshold value.
The first step：By the image gray processing of acquisition, the gray value of each pixel of acquisition image.
Second step：The position x of random initializtion crow groupi,j(representing a gray value), according to the k values of input, by image It is divided into k threshold (j=1,2 ..., k), with maximum entropy method (MEM), obtains the maximum entropy f of k threshold.
3rd step；According to the principle of crow group's algorithm, crow, which is known, to be followed, according to algorithm
Iterative formula (1), carry out position iteration and obtain new position x 'i,j, bring the value newly obtained into maximum entropy method (MEM) and try to achieve New entropy f', compares f and f' size, maximum is preserved.
4th step:Crow, which is not known, oneself to be followed, and according to the iterative formula (2) of algorithm, is carried out position iteration and is obtained newly Position x "i,j, bring the value newly obtained into maximum entropy method (MEM) and try to achieve new entropy f ", compare f and f " size, maximum is protected Leave and.
5th step：The segmentation result x of the k threshold value obtained according to maximum entropyi,j(j=1,2 ... it is exactly k) to be determined Segmentation threshold.
Segmentation figure picture：Suitable rear progress image segmentation is determined based on the 5th step judgment threshold, two face are now presented on image Color we purpose signature grey scale value can be set to 0 be black, other positions are uniformly set gray value 255 be white.
Extract part interested (sugarcane section information)：Based on the completion of the 6th step segmentation figure picture, image shows two kinds of black and white Color, black are our parts interested (sugarcane section).
Accuracy rate detects：When completing 7 step, the process that identifies substantially is completed, we can by artificial detection come Judge whether the position of identification is accurate：If accurate, the 9th step is carried out；If inaccurate, the 5th step is returned.
It is stored in database：Threshold value used deposit database after 8th step is identified successfully, constantly long term accumulation, expansion, rich Rich database.This threshold value is arranged to Ri, reference threshold is R in databaseoSo Ri-Ro>=M, M are what lot of experimental data was drawn Error.If meet condition RiStorage, on the contrary give up.
Determine sugarcane section position：After being come out by threshold method to each sugarcane sugarcane section Automatic feature recognition, pass through the picture of sampling Vegetarian refreshments obtains the size positions of sugarcane sugarcane section and the specific length respectively saved, and input data compared with the length dimension demarcated Storehouse, and it is supplied to kind of machine of cutting to carry out cutting kind of an operation.
This method takes crow colony intelligence optimized algorithm to improve the intelligent decision to image threshold, and accurate segmentation figure picture carries The position of sugarcane section is taken out, and establishes database and experimental result data is stored in database analysis, to improve the accuracy rate of identification.This Goal of the invention is to realize that rapidly and accurately sugarcane sugarcane section position is identified, and cuts the automatic of kind of machine for follow-up intelligence and cuts kind of an operation Technical support is provided.
The preferred embodiment to the invention is illustrated above, but the present invention is not limited to embodiment, Those skilled in the art can also make a variety of equivalent modifications on the premise of without prejudice to the invention spirit Or replace, these equivalent modifications or replacement are all contained in scope of the present application.

Claims (6)

1. a kind of recognition methods of sugarcane sugarcane section feature, it is characterised in that comprise the following steps：
Step 1：The sugarcane for choosing different cultivars carries out IMAQ, obtains gathering image；
Step 2：Denoising is carried out to collection image and gray processing handles to obtain pretreatment image；
Step 3：European cluster analysis is carried out to pretreatment image to obtain analyzing image；
Step 4：Gray value average computation is carried out to analysis image；
Step 5：Computing is carried out to the image after calculating gray value using crow colony intelligence optimized algorithm and determines sugarcane sugarcane section Characteristic threshold value；
Step 6：Dividing processing is carried out to image according to the characteristic threshold value of the sugarcane sugarcane section in step 5；
Step 7：Sugarcane sugarcane section information extraction is carried out to the image after segmentation；
Step 8：The sugarcane sugarcane section information of extraction is accurately detected, if accurately, into next step；If inaccurate, Return to the 5th step；
Step 9：It is R reference threshold in threshold value and databaseoCarry out computing, Ri-Ro>=M, M are what lot of experimental data was drawn Error, RiFor threshold value, if meeting that condition is put in storage, otherwise give up.
A kind of 2. recognition methods of sugarcane sugarcane section feature according to claim 1, it is characterised in that：Also include passing through threshold value After method comes out to each sugarcane sugarcane section Automatic feature recognition, by the pixel of sampling compared with the length dimension demarcated, ask Go out the size positions of sugarcane sugarcane section and the specific length respectively saved, and input database, and be supplied to kind of machine of cutting to carry out cutting kind of an operation.
A kind of 3. recognition methods of sugarcane sugarcane section feature according to claim 1, it is characterised in that：It is right in the step 1 The process of sugarcane IMAQ is to carry out IMAQ to the shape and color of variety classes sugarcane.
A kind of 4. recognition methods of sugarcane sugarcane section feature according to claim 1, it is characterised in that：In the step 5 really The detailed process for determining the characteristic threshold value of sugarcane sugarcane section is：Using image one of those or N number of pixel as a particle, adopt Computing, which is done, by the use of crow colony intelligence optimized algorithm finds an intensity value ranges as the characteristic threshold value of the sugarcane sugarcane section of reference, and Judge whether this threshold value is suitable, if appropriate then threshold value, into next step, if improper return to step 4 recalculates Find threshold value.
A kind of 5. recognition methods of sugarcane sugarcane section feature according to claim 1, it is characterised in that：Scheme in the step 6 As the detailed process of segmentation is：It is black that purpose sign gray value is set into 0, and other positions are uniformly set into gray value 255 i.e. For white, black and white segmentation is carried out to image according to threshold values.
A kind of 6. recognition methods of sugarcane sugarcane section feature according to claim 1, it is characterised in that：It is sweet in the step 7 The process of sugarcane sugarcane section information extraction is：Information extraction is carried out to picture black part.
CN201711061414.5A 2017-11-02 2017-11-02 A kind of recognition methods of sugarcane sugarcane section feature Pending CN107680098A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711061414.5A CN107680098A (en) 2017-11-02 2017-11-02 A kind of recognition methods of sugarcane sugarcane section feature

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711061414.5A CN107680098A (en) 2017-11-02 2017-11-02 A kind of recognition methods of sugarcane sugarcane section feature

Publications (1)

Publication Number Publication Date
CN107680098A true CN107680098A (en) 2018-02-09

Family

ID=61144757

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711061414.5A Pending CN107680098A (en) 2017-11-02 2017-11-02 A kind of recognition methods of sugarcane sugarcane section feature

Country Status (1)

Country Link
CN (1) CN107680098A (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108875789A (en) * 2018-05-23 2018-11-23 广西民族大学 A kind of sugarcane sugarcane bud specific identification device based on deep learning
CN108876767A (en) * 2018-05-23 2018-11-23 广西民族大学 A kind of quick identification device of sugarcane sugarcane section feature
CN108960100A (en) * 2018-06-22 2018-12-07 广西大学 A kind of recognition methods of the sugarcane sugarcane section based on image procossing
CN109168499A (en) * 2018-09-30 2019-01-11 广西民族大学 A kind of sugarcane pre-cut kind work station
CN112673777A (en) * 2020-12-22 2021-04-20 钦州市农业科学研究所 Automatic seed descending system for sugarcane for agricultural production and control method
CN113916172A (en) * 2021-10-21 2022-01-11 柳州市孚桂智能科技有限公司 Sugarcane festival position detection device
CN114119944A (en) * 2021-12-23 2022-03-01 天津天地伟业信息系统集成有限公司 Image detection device, method and computer readable storage medium

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102663681A (en) * 2012-03-11 2012-09-12 西安电子科技大学 Gray scale image segmentation method based on sequencing K-mean algorithm
CN102663434A (en) * 2012-04-24 2012-09-12 广西大学 Cane stalk recognition method
CN102938052A (en) * 2011-08-16 2013-02-20 汪建 Sugarcane segmentation and recognition method based on computer vision
CN105225228A (en) * 2015-09-08 2016-01-06 广西大学 Leifsonia image partition method under the natural background of field
CN105279244A (en) * 2015-09-30 2016-01-27 广西大学 Method for establishing sugarcane seed bud classification feature database
CN105654099A (en) * 2014-08-25 2016-06-08 崔胡晋 Sugarcane segmentation and identification method based on improved vision
CN105761237A (en) * 2015-12-15 2016-07-13 江南大学 Mean shift-based chip X-ray image layer segmentation

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102938052A (en) * 2011-08-16 2013-02-20 汪建 Sugarcane segmentation and recognition method based on computer vision
CN102663681A (en) * 2012-03-11 2012-09-12 西安电子科技大学 Gray scale image segmentation method based on sequencing K-mean algorithm
CN102663434A (en) * 2012-04-24 2012-09-12 广西大学 Cane stalk recognition method
CN105654099A (en) * 2014-08-25 2016-06-08 崔胡晋 Sugarcane segmentation and identification method based on improved vision
CN105225228A (en) * 2015-09-08 2016-01-06 广西大学 Leifsonia image partition method under the natural background of field
CN105279244A (en) * 2015-09-30 2016-01-27 广西大学 Method for establishing sugarcane seed bud classification feature database
CN105761237A (en) * 2015-12-15 2016-07-13 江南大学 Mean shift-based chip X-ray image layer segmentation

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
ALIREZA ASKARZADEH: "A novel metaheuristic method for solving constrained engineering optimization problems: Crow search algorithm", 《COMPUTERS AND STRUCTURES 》 *
DIEGO OLIVA ET AL.: "Cross entropy based thresholding for magnetic resonance brain images using Crow Search Algorithm", 《EXPERT SYSTEMS WITH APPLICATIONS》 *
K.MOSHASHAI ET AL.: "Identification of Sugarcane Nodes Using Image Processing and Machine Vision Technology", 《INTERNATIONAL JOURNAL OF AGRICULTURAL RESEARCH》 *
张卫正 等: "基于图像处理的甘蔗茎节识别与定位", 《农机化研究》 *
韦相贵: "基于智能算法的甘蔗定位切割方法", 《江苏农业科学》 *

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108875789A (en) * 2018-05-23 2018-11-23 广西民族大学 A kind of sugarcane sugarcane bud specific identification device based on deep learning
CN108876767A (en) * 2018-05-23 2018-11-23 广西民族大学 A kind of quick identification device of sugarcane sugarcane section feature
CN108875789B (en) * 2018-05-23 2021-04-27 广西民族大学 Sugarcane bud feature recognition device based on deep learning
CN108876767B (en) * 2018-05-23 2021-04-27 广西民族大学 Sugarcane festival characteristic quick identification device
CN108960100A (en) * 2018-06-22 2018-12-07 广西大学 A kind of recognition methods of the sugarcane sugarcane section based on image procossing
CN109168499A (en) * 2018-09-30 2019-01-11 广西民族大学 A kind of sugarcane pre-cut kind work station
CN109168499B (en) * 2018-09-30 2023-07-18 广西民族大学 Sugarcane precut seed workstation
CN112673777A (en) * 2020-12-22 2021-04-20 钦州市农业科学研究所 Automatic seed descending system for sugarcane for agricultural production and control method
CN113916172A (en) * 2021-10-21 2022-01-11 柳州市孚桂智能科技有限公司 Sugarcane festival position detection device
CN113916172B (en) * 2021-10-21 2023-09-01 柳州市孚桂智能科技有限公司 Sugarcane section position detection device
CN114119944A (en) * 2021-12-23 2022-03-01 天津天地伟业信息系统集成有限公司 Image detection device, method and computer readable storage medium

Similar Documents

Publication Publication Date Title
CN107680098A (en) A kind of recognition methods of sugarcane sugarcane section feature
CN111709379A (en) Remote sensing image-based hilly area citrus planting land plot monitoring method and system
Tian et al. Application status and challenges of machine vision in plant factory—A review
CN107977671A (en) A kind of tongue picture sorting technique based on multitask convolutional neural networks
Alencastre-Miranda et al. Robotics for sugarcane cultivation: Analysis of billet quality using computer vision
CN106355143A (en) Seed maize field identification method and system based on multi-source and multi-temporal high resolution remote sensing data
Ok et al. 2-D delineation of individual citrus trees from UAV-based dense photogrammetric surface models
CN113657158B (en) Google EARTH ENGINE-based large-scale soybean planting area extraction algorithm
CN115861629A (en) High-resolution farmland image extraction method
CN113221806A (en) Cloud platform fusion multi-source satellite image and tea tree phenological period based automatic tea garden identification method
CN111882573B (en) Cultivated land block extraction method and system based on high-resolution image data
CN116129260A (en) Forage grass image recognition method based on deep learning
CN114627380A (en) Rice identification method based on fusion of optical image and SAR time sequence data
CN112036313A (en) Tobacco planting area detection method, device and equipment and readable storage medium
CN115861686A (en) Litchi key growth period identification and detection method and system based on edge deep learning
Mathews Applying geospatial tools and techniques to viticulture
Paulo et al. Wheat lodging ratio detection based on UAS imagery coupled with different machine learning and deep learning algorithms
Matias et al. Bison‐Fly: An open‐source UAV pipeline for plant breeding data collection
CN116052141B (en) Crop growth period identification method, device, equipment and medium
CN117152609A (en) Crop appearance characteristic detecting system
CN105046229B (en) A kind of recognition methods of crops row and device
CN116524344A (en) Tomato string picking point detection method based on RGB-D information fusion
Li et al. Semi-supervised cooperative regression model for small sample estimation of citrus leaf nitrogen content with UAV images
Li et al. A longan yield estimation approach based on uav images and deep learning
CN114170518A (en) Tea tree freezing injury assessment method and system based on computer vision

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20180209