CN104200193A - Fruit tree yield estimation method and device - Google Patents
Fruit tree yield estimation method and device Download PDFInfo
- Publication number
- CN104200193A CN104200193A CN201410381889.2A CN201410381889A CN104200193A CN 104200193 A CN104200193 A CN 104200193A CN 201410381889 A CN201410381889 A CN 201410381889A CN 104200193 A CN104200193 A CN 104200193A
- Authority
- CN
- China
- Prior art keywords
- fruit
- image
- fruit tree
- tree
- circle
- 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)
- Image Processing (AREA)
Abstract
The invention relates to the field of agricultural information technology, and discloses a fruit tree yield estimation method and a fruit tree yield estimation device. The method comprises the steps of: S101, obtaining a crown image of a single fruit tree in a mature stage and position data of the fruit tree; S102, identifying number of fruits in the image according to color and shape characteristics of the fruits; S103, computing fruit yield data of the single fruit tree; S104, generating a yield statistical chart according to the position data of the fruit tree and the computed fruit yield data of the single fruit tree. By using the fruit tree yield estimation method and device provided by the invention, professional devices and technical staffs are not needed, method is simple, economic cost is low, repeatability is high; furthermore, the method focuses on the single fruit tree instead of a whole planting area, thus being more suitable for scattered farmers, and convenient for the farmers to rapidly adjust management strategies according to actual situations.
Description
Technical field
The present invention relates to Agricultural Information technical field, particularly a kind of output of the fruit tree estimating and measuring method and device.
Background technology
Chinese apple cultivated area and the output Jun Ju world first, also be maximum apple country of consumption simultaneously, but orchard management level is relatively backward, the more restriction that is subject to labor management level and natural climate, there is very large instability in therefore annual apple production, and further causes the fluctuation of market supply and orchard worker's income.If plantation family can be estimated apple production by some relatively simple methods in apple development process, and adjust in time production management and sales tactics according to predictive output, can effectively save the means of production and drop into, and improve plantation family income, reach the doulbe-sides' victory of economy and zoology benefit.Yet two kinds of at present relevant to recovery prediction main stream approach are all more complicated, and a kind of is to utilize remote sensing technology, by image spectral information, analyze the parameters such as NVDI, the crop yield of prediction large area scope; Another kind is by professional biochemistry detection equipment, detects the factors relevant to output such as crops physical signs, growing environment, and the relation between the analysis of biochemical factor and output, sets up Production Forecast Models.
Current existing main stream approach, highly professional, and use cost is higher.Remote sensing technology estimation output needs the high spectrum image in output estimation region, by analyzing the biophysical parameters of inverting crops, more be applied to large area proportion of crop planting region in flakes, and price image is expensive, is applicable to the macro-level policy-making of government administration section; And forecast model method depends on a series of biochemical parameters such as crops physical signs, envirment factor, parameter acquiring needs professional measuring equipment, and zones of different has different parameter values, and each prediction all needs resampling chemical examination, obtains mode input value.Therefore, existing main stream approach all needs professional technician's guidance and the support of professional equipment, and output estimation material requested needs overlapping investment, cost is also relatively high, and complex operation step, predetermined period is relatively long, is unfavorable for applying in common plantation family.
Summary of the invention
For addressing the above problem; the present invention proposes a kind of output of the fruit tree estimating and measuring method and device; while adopting the method and device to carry out output of the fruit tree estimation; do not need professional equipment and technician; simple to operate; repeatable high, and predetermined period is shorter, when making decision-making of production management, the planting fruit trees family that is applicable to non-scale uses.
The present invention proposes a kind of output of the fruit tree estimating and measuring method, said method comprising the steps of:
S101, the tree crown image that obtains maturity stage individual plant fruit tree and the position data of fruit tree;
S102, according to the CF feature of fruit, identify the number of fruit in described image;
S103, according to recognition result, calculate the fruit yield data of described individual plant fruit tree;
S104, the described fruit yield data generation output statistics figure according to the position data of described fruit tree with the individual plant fruit tree calculating.
Preferably, described step S102 specifically comprises:
According to the color characteristic of fruit, be provided for distinguishing the pixel segmentation threshold value of fruit and tree body in described image;
According to described pixel segmentation threshold value, described image is carried out to binary conversion treatment, obtain characterizing the binary image of fruit and tree body;
Described binary image is carried out to noise removal process;
According to the shape facility of fruit, the image after noise remove is justified to matching;
Extract the quantity of the circle generating in the image after circle matching, obtain the number of fruit in described image.
Preferably, describedly according to the shape facility of fruit, the image after noise remove is justified to matching and specifically comprises:
By edge detection algorithm, in binary image, extract the fruit edge line of all fruits;
Select arbitrary fruit edge line, certain that determine described fruit edge line is some edge curvature zequin;
Along described fruit edge line, with fixed step size, carry out continuous sampling, calculate the curvature of each sampled point;
Whether the curvature that judges sampled point meets default rim condition, if do not met, reselects starting point and carries out curvature calculating, as met, described fruit edge line matching is generated to circle, obtains matching radius of a circle and the center of circle of generation;
Judge that any two matchings generate the center of circle of circle and radius difference whether within the scope of error threshold, if not, be defined as two circles, if so, will described two matchings generate to justify and merge into a circle.
Preferably, described default rim condition comprises:
The absolute value of single sampled point curvature is less than C
maxand be greater than C
min; The curvature difference absolute value of neighbouring sample point is less than maximum tolerance C
dif; The sampling number that meets continuously above-mentioned two conditions is more than or equal to m;
Wherein, C
maxand C
minmaximal value and the minimum value of the normal value of imaging apple curvature in image, C
difto block with overlapping in the situation that, while there is irregular sudden change in edge curvature, the MAD of curvature value variation; C
max, C
min, C
difbe with m the value that analysis obtains through great many of experiments.
Preferably, by described image, the R/B value in RGB image model and the V value of described image in HSV image model form described pixel segmentation threshold value.
Preferably, described step S104 specifically comprises:
The position data of described fruit tree is combined with the apple production data of the individual plant fruit tree estimating, generates the yield data of individual plant fruit tree;
Add up the yield data of the individual plant fruit tree of all fruit trees, generate and take individual plant fruit tree as individual output statistics figure.
The invention allows for a kind of output of the fruit tree estimating apparatus of applying method described in above-mentioned any one, described device comprises equipment body and controller, described equipment body comprises GPS receiver module and high-definition camera module, and described controller comprises image processing module, computing module, generation display module and control module;
GPS receiver module, for gathering the position data of maturity stage individual plant fruit tree;
High-definition camera module, for taking the tree crown image of maturity stage individual plant fruit tree;
Image processing module, for identifying the number of described image fruit according to the CF feature of fruit;
Computing module, calculates the fruit yield data of described individual plant fruit tree for the number of the described image fruit that identifies according to image processing module;
Generate display module, for generating output statistics figure according to the position data of described fruit tree and the described fruit yield data of the individual plant fruit tree calculating;
Control module, for input control information, controls modules.
Preferably, described GPS receiver module is rotatable, for making gps antenna all the time perpendicular to surface level.
Preferably, described image processing module specifically comprises:
Segmentation threshold setting unit, for being provided for distinguishing at described image the pixel segmentation threshold value of fruit and tree body according to the color characteristic of fruit;
Binary conversion treatment unit, for described image being carried out to binary conversion treatment according to described pixel segmentation threshold value, obtains characterizing the binary image of fruit and tree body;
Denoising unit, for carrying out noise removal process to described binary image;
Circle matching unit, for justifying matching according to the shape facility of fruit by the image after noise remove;
Extraction unit, the quantity of the circle generating for the image extracting after round matching, obtains the number of fruit in described image.
Preferably, described equipment body is connected by data cable with controller.
The present invention proposes a kind of output of the fruit tree estimating and measuring method and device, by the number of apple in the method identification photographic images of image recognition, and then estimate the output of this individual plant fruit tree, and yield data is combined with fruit tree position data generate output statistics figure, realize output of the fruit tree estimation.The method does not need professional equipment and technician, and method is simple, repeatable high; And the method is from microcosmic angle, it is individual that emphasis is conceived to individual plant fruit tree, and no longer from whole planting area angle, be more suitable for scattered plantation family, is convenient to plant family according to fruit tree actual conditions rapid adjustment operating strategy, and specific aim is stronger.
Accompanying drawing explanation 3
Fig. 1 is the process flow diagram of a kind of output of the fruit tree estimating and measuring method of proposing of the present invention;
Fig. 2 is the process flow diagram of the fruit identification that proposes in the embodiment of the present invention;
Fig. 3 is the process flow diagram of the round Fitting Analysis that proposes in the embodiment of the present invention;
Fig. 4 is the structural representation of a kind of output of the fruit tree estimating apparatus of proposing of the present invention;
Fig. 5 is the front elevation of the equipment body that proposes in the embodiment of the present invention;
Fig. 6 is the side view of the equipment body that proposes in the embodiment of the present invention.
Embodiment
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used for illustrating the present invention, but are not used for limiting the scope of the invention.
Fig. 1 is the process flow diagram of a kind of output of the fruit tree estimating and measuring method of the embodiment of the present invention one proposition, as shown in Figure 1, said method comprising the steps of:
S101, the tree crown image that obtains maturity stage individual plant fruit tree and the position data of fruit tree;
S102, according to the CF feature of fruit, identify the number of fruit in described image;
S103, according to recognition result, calculate the fruit yield data of described individual plant fruit tree;
S104, the described fruit yield data generation output statistics figure according to the position data of described fruit tree with the individual plant fruit tree calculating.
The output of the fruit tree estimating and measuring method that the present invention proposes is realized based on image pattern recognition, and in the method, the flow process of step S102 fruit identification, as shown in Figure 2, specifically comprises:
S201, the pixel segmentation threshold value that is provided for distinguishing fruit in described image and sets body according to the color characteristic of fruit; Wherein, by described image, the R/B value in RGB image model and the V value of described image in HSV image model form described pixel segmentation threshold value.
S202, according to described pixel segmentation threshold value, described image is carried out to binary conversion treatment, obtain characterizing the binary image of fruit and tree body;
S203, described binary image is carried out to noise removal process;
S204, according to the shape facility of fruit, the image after noise remove is justified to matching;
S205, extract the quantity of the circle generating in the image after circle matching, obtain the number of fruit in described image.
The embodiment of the present invention, the apple tree of take is specifically described as example, but is not used for limiting the scope of the invention.
In embodiments of the present invention, the image of equipment body passback can show in the display screen of controller, then in image, draws at random a horizontal line from image left side to right side, and user also can adjust horizontal line position up and down according to picture material.After determining that above operation is errorless, user is by clicking display module, from left to right successively the central point of all apples of above horizontal line process is chosen, method calculates R/B value (RGB image model) and the V value (HSV image model) of all user's selected element apples subsequently, and counting maximal value and minimum value, user can select the required threshold value of apple identification within the scope of this.
According to previous experiments result, in image, the pixel R/B of apple position and V value generally there will be peak value (maximal value), there is notable difference with other pixel R/B value and V value, therefore can be between maximal value and minimum value selected threshold, as the pixel segmentation threshold value of distinguishing apple and tree body other parts.According to the segmentation threshold obtaining in above step, the method is carried out binary conversion treatment to image, pixel value higher than threshold value is 1 (white), lower than the pixel value of threshold value, is 0 (black), original image is converted into the binary image of black background white print spot.Image, after cutting apart, because fruit tree other parts R/B value and V value in image may be greater than got threshold value, also can present irregular hickie in binary image, but generally all not too large.By morphologic filtering method, the medium and small figure spot of image is carried out to erosion operation, merge with black background, can reach the object of eliminating noise.
After above processing, in image, hickie quantity has been similar to the upper apple quantity of tree.Because the apple in three dimensions projects in image, can produce and block and overlapping phenomenon, therefore can produce the phenomenon that an apple identification is become to two hickies or two apple identifications are become to a hickie, need to justify Fitting Analysis, reduce the error between the true quantity of hickie quantity and fruit.After circle Fitting Analysis, can avoid overlapping and count and many meter phenomenons with blocking produced apple quantity leakage, then the round quantity of generation in statistical picture, the i.e. upper apple quantity of tree.According to single apple average weight, estimate this strain output of the fruit tree, be combined and can generate an output record with GPS position data, can be used for the later stage by fruit tree generation output statistics figure.
After the upper apple of tree is in projecting to image, there will be overlapping and phenomenon that block, the key addressing this problem is circle fitting technique.Conventionally in image, apple edge can present regular circular shape, and be blocked or when overlapping, circular arc there will be the irregular variations such as depression, projection, by calculating hickie edge curvature in binary image, can revise counting error, as shown in Figure 3, described step S201 justifies matching according to the shape facility of fruit by the image after noise remove, specifically comprises:
S301, by edge detection algorithm, in binary image, extract the fruit edge line of all fruits;
S302, select arbitrary fruit edge line, certain that determine described fruit edge line is some edge curvature zequin;
S303, along described fruit edge line, with fixed step size, carry out continuous sampling, calculate the curvature of each sampled point;
S304, judge that whether the curvature of sampled point meet default rim condition, if do not met, return to step S302, reselect starting point and carry out curvature calculating, as met, perform step S305;
S305, described fruit edge line matching is generated to circle, obtain matching radius of a circle and the center of circle of generation;
S306, judge that any two matchings generate the center of circle of circle and radius difference whether within the scope of error threshold, if not, perform step S307, generates two circles, if so, perform step S308, will described two matchings generation justify and merge into a circle.
In embodiments of the present invention, the binary image after cutting apart based on image, by Canny edge detection operator by apple edge line extraction potential in image out, the more any edge line of random selection, from arbitrfary point, start edge calculation curvature.Along edge line, preferably step-length is that 5 pixels carry out continuous sampling, also can set other step values according to practical application, calculate the curvature of each sample point, whether the curvature of analyzing again each sample point meets rim condition, as met, assert it is an apple edge, otherwise reselect starting point, carry out curvature analysis.Rim condition comprises following several: 1. the absolute value of single sampled point curvature is less than C
maxand be greater than C
min, be the normal span of imaging apple curvature in image.2. the curvature difference absolute value of neighbouring sample point is less than maximum tolerance C
dif.Blocking with overlapping in the situation that, edge curvature is easy to occur irregular sudden change, thereby causes the larger variation of curvature value.3. satisfy condition continuously 1 and the sampling number of condition 2 be more than or equal to m.C in above condition
max, C
min, C
difwith m be all the empirical value that great many of experiments analysis obtains, different according to apple variety, value is difference slightly.
Edge fitting after all processing is generated to circle, obtain and generate matching radius of a circle and the center of circle, compare matching radius of a circle and the center of circle, as any two matchings generate the center of circle spacing of circle and radius difference in the allowed band of error threshold, circle is merged, be defined as single circle, be single apple, otherwise generate two circles, repeatedly generate more afterwards a plurality of circles, i.e. a plurality of apples.
Further, described step S104 specifically comprises:
The position data of described fruit tree is combined with the apple production data of the individual plant fruit tree estimating, generates the yield data of individual plant fruit tree;
Add up the yield data of the individual plant fruit tree of all fruit trees, generate and take individual plant fruit tree as individual output statistics figure, take fruit tree as unit generates output statistics figure, further, estimate the apple production level in whole orchard.
The output of the fruit tree estimating and measuring method that the present invention proposes, R/B value in Integrated using RGB pattern and the V value in HSV pattern, the segmentation threshold that apple is different from fruit tree other parts is obtained in analysis, cuts apart image to generate binary image, and in this image, figure spot quantity is similar to apple quantity.Based on apple uniform characteristic of projection back edge curvature in image, by analyzing the ANOMALOUS VARIATIONS of apple edge curvature, reject because blocking and overlapping generated non-boundary curve, according to boundary curve, generate fitting circle again, by analyzing fitting circle, reduce the counting error of apple identification, thereby add up exactly apple quantity, estimate apple total production.
Another embodiment of the present invention has also proposed a kind of output of the fruit tree estimating apparatus of applying said method, as shown in Figure 4, described device comprises equipment body 401 and controller 402, described equipment body 401 comprises GPS receiver module 4011 and high-definition camera module 4012, and described controller 402 comprises image processing module 4021, computing module 4022, generates display module 4023 and control module 4024;
GPS receiver module 4011, for gathering the position data of maturity stage individual plant fruit tree;
High-definition camera module 4012, for taking the tree crown image of maturity stage individual plant fruit tree;
Image processing module 4021, for identifying the number of described image fruit according to the CF feature of fruit;
Computing module 4022, calculates the fruit yield data of described individual plant fruit tree for the number of the described image fruit that identifies according to image processing module;
Generate display module 4023, for generating output statistics figure according to the position data of described fruit tree and the described fruit yield data of the individual plant fruit tree calculating;
Control module 4024, for input control information, controls modules.
The equipment that the output of the fruit tree estimation relating in the embodiment of the present invention relies on, comprise two parts, a part is equipment body, during use, set up Huo Zheng side, fruit tree top, at least should there is GPS receiver module, high-definition camera module, as shown in Fig. 5 and Fig. 6, wherein GPS receiver module is rotatable, can make gps antenna all the time perpendicular to surface level; Another part is to set up higher in order to solve equipment body, being difficult for direct control problem designs, the controller of taking, locating, process in order to opertaing device main body, at least comprise image processing module, computing module, generation display module and control module, image processing module wherein, computing module and control module are positioned at controller inside.Controller is connected by data cable with equipment body, swap data.
Tree type below by apple tree in test orchard be take spindle as example, the present invention is specifically described, before using, equipment body being erected to the position that can cover whole tree crown is tree crown top, photographing module is towards tree crown, height and the angle of by controller display module, adjusting equipment body, can cover whole tree crown.Hedge wall type tree body equipment for other tree-like fruit tree particular locations, by fruit tree tree type, determined, as need to be placed on and the parallel plane position of hedge wall.After equipment body position is fixing, by the high-definition camera module of controller starting outfit main body, open GPS location simultaneously.Equipment body photographic images also passes back on the generation display module of controller, on image, can automatically generate a horizontal line for threshold calculations.As required, user also can move up and down this horizontal line, makes horizontal line can pass through apple and fruit tree other parts simultaneously.Complete after above operation, user will, all apples of horizontal line process, by clicking screen, choose successively.Now, controller can calculate the R/B of selected apple and the span of V value automatically, provides threshold range and selects for user, and user selects pixel segmentation threshold value, and selects preset C
max, C
min, C
difafter m parameter value, system can be carried out the estimation of output automatically, and the positional information of having located and the upper apple quantity information of tree are shown simultaneously.User also can, by changing above parameter, re-start estimation to output.Preserve above information, complete the yield by estimation of this strain fruit tree, repeat above operation and can estimate other output of the fruit tree.
Preferably, described image processing module 4021 specifically comprises:
Segmentation threshold setting unit, for being provided for distinguishing at described image the pixel segmentation threshold value of fruit and tree body according to the color characteristic of fruit;
Binary conversion treatment unit, for described image being carried out to binary conversion treatment according to described pixel segmentation threshold value, obtains characterizing the binary image of fruit and tree body;
Denoising unit, for carrying out noise removal process to described binary image;
Circle matching unit, for justifying matching according to the shape facility of fruit by the image after noise remove;
Extraction unit, the quantity of the circle generating for the image extracting after round matching, obtains the number of fruit in described image.
Preferably, described equipment body 401 is connected by data cable with controller 402.
The device that the present embodiment proposes does not need professional equipment, simple to operate, without professional and technical personnel, can operate; output estimation efficiency is higher; can immediately obtain recovery prediction value, and the financial cost of equipment is lower, when decision-making of production management is made at the planting fruit trees family that is applicable to non-scale, uses.
A kind of output of the fruit tree estimating and measuring method and the device that adopt the present invention to propose, CF feature by maturity stage apple extracts apple from background image, and estimates fast side individual plant output of the fruit tree according to apple quantity, in conjunction with GPS position data, can also generate output of the fruit tree statistical graph.Not only utilization threshold is lower in the present invention, and predetermined period is shorter, when decision-making of production management is made at the planting fruit trees family that is applicable to non-scale, uses, and the method is also a kind of new trial to output estimation simultaneously.
Through the above description of the embodiments, those skilled in the art can be well understood to the present invention and can realize by hardware, and the mode that also can add necessary general hardware platform by software realizes.Understanding based on such, technical scheme of the present invention can embody with the form of software product, it (can be CD-ROM that this software product can be stored in a non-volatile memory medium, USB flash disk, portable hard drive etc.) in, comprise some instructions with so that computer equipment (can be personal computer, server, or the network equipment etc.) carry out the method described in each embodiment of the present invention.
It will be appreciated by those skilled in the art that accompanying drawing is the schematic diagram of a preferred embodiment, the module in accompanying drawing or flow process might not be that enforcement the present invention is necessary.
It will be appreciated by those skilled in the art that the module in the device in embodiment can be distributed in the device of embodiment according to embodiment description, also can carry out respective change and be arranged in the one or more devices that are different from the present embodiment.The module of above-described embodiment can be merged into a module, also can further split into a plurality of submodules.
Disclosed is above only several specific embodiment of the present invention, and still, the present invention is not limited thereto, and the changes that any person skilled in the art can think of all should fall into protection scope of the present invention.
Claims (10)
1. an output of the fruit tree estimating and measuring method, is characterized in that, said method comprising the steps of:
S101, the tree crown image that obtains maturity stage individual plant fruit tree and the position data of fruit tree;
S102, according to the CF feature of fruit, identify the number of fruit in described image;
S103, according to recognition result, calculate the fruit yield data of described individual plant fruit tree;
S104, the described fruit yield data generation output statistics figure according to the position data of described fruit tree with the individual plant fruit tree calculating.
2. the method for claim 1, is characterized in that, described step S102 specifically comprises:
According to the color characteristic of fruit, be provided for distinguishing the pixel segmentation threshold value of fruit and tree body in described image;
According to described pixel segmentation threshold value, described image is carried out to binary conversion treatment, obtain characterizing the binary image of fruit and tree body;
Described binary image is carried out to noise removal process;
According to the shape facility of fruit, the image after noise remove is justified to matching;
Extract the quantity of the circle generating in the image after circle matching, obtain the number of fruit in described image.
3. method as claimed in claim 2, is characterized in that, describedly according to the shape facility of fruit, the image after noise remove is justified to matching and specifically comprises:
By edge detection algorithm, in binary image, extract the fruit edge line of all fruits;
Select arbitrary fruit edge line, certain that determine described fruit edge line is some edge curvature zequin;
Along described fruit edge line, with fixed step size, carry out continuous sampling, calculate the curvature of each sampled point;
Whether the curvature that judges sampled point meets default rim condition, if do not met, reselects starting point and carries out curvature calculating, as met, described fruit edge line matching is generated to circle, obtains matching radius of a circle and the center of circle of generation;
Judge that any two matchings generate the center of circle of circle and radius difference whether within the scope of error threshold, if not, be defined as two circles, if so, will described two matchings generate to justify and merge into a circle.
4. method as claimed in claim 3, is characterized in that, described default rim condition comprises:
The absolute value of single sampled point curvature is less than C
maxand be greater than C
min; The curvature difference absolute value of neighbouring sample point is less than maximum tolerance C
dif; The sampling number that meets continuously above-mentioned two conditions is more than or equal to m;
Wherein, C
maxand C
minmaximal value and the minimum value of the normal value of imaging apple curvature in image, C
difto block with overlapping in the situation that, while there is irregular sudden change in edge curvature, the MAD of curvature value variation; C
max, C
min, C
difbe with m the value that analysis obtains through great many of experiments.
5. method as claimed in claim 2, is characterized in that, by described image, the R/B value in RGB image model and the V value of described image in HSV image model form described pixel segmentation threshold value.
6. the method for claim 1, is characterized in that, described step S104 specifically comprises:
The position data of described fruit tree is combined with the apple production data of the individual plant fruit tree estimating, generates the yield data of individual plant fruit tree;
Add up the yield data of the individual plant fruit tree of all fruit trees, generate and take individual plant fruit tree as individual output statistics figure.
7. an application rights requires the output of the fruit tree estimating apparatus of method described in 1 to 6 any one, it is characterized in that, described device comprises equipment body and controller, described equipment body comprises GPS receiver module and high-definition camera module, and described controller comprises image processing module, computing module, generation display module and control module;
GPS receiver module, for gathering the position data of maturity stage individual plant fruit tree;
High-definition camera module, for taking the tree crown image of maturity stage individual plant fruit tree;
Image processing module, for identifying the number of described image fruit according to the CF feature of fruit;
Computing module, calculates the fruit yield data of described individual plant fruit tree for the number of the described image fruit that identifies according to image processing module;
Generate display module, for generating output statistics figure according to the position data of described fruit tree and the described fruit yield data of the individual plant fruit tree calculating;
Control module, for input control information, controls modules.
8. device as claimed in claim 7, is characterized in that, described GPS receiver module is rotatable, for making gps antenna all the time perpendicular to surface level.
9. device as claimed in claim 7, is characterized in that, described image processing module specifically comprises:
Segmentation threshold setting unit, for being provided for distinguishing at described image the pixel segmentation threshold value of fruit and tree body according to the color characteristic of fruit;
Binary conversion treatment unit, for described image being carried out to binary conversion treatment according to described pixel segmentation threshold value, obtains characterizing the binary image of fruit and tree body;
Denoising unit, for carrying out noise removal process to described binary image;
Circle matching unit, for justifying matching according to the shape facility of fruit by the image after noise remove;
Extraction unit, the quantity of the circle generating for the image extracting after round matching, obtains the number of fruit in described image.
10. system as claimed in claim 7, is characterized in that, described equipment body is connected by data cable with controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410381889.2A CN104200193A (en) | 2014-08-05 | 2014-08-05 | Fruit tree yield estimation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410381889.2A CN104200193A (en) | 2014-08-05 | 2014-08-05 | Fruit tree yield estimation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104200193A true CN104200193A (en) | 2014-12-10 |
Family
ID=52085483
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410381889.2A Pending CN104200193A (en) | 2014-08-05 | 2014-08-05 | Fruit tree yield estimation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104200193A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104616004A (en) * | 2015-03-08 | 2015-05-13 | 无锡桑尼安科技有限公司 | Citrus yield estimation method based on multi-estimation parameters |
| CN104636722A (en) * | 2015-01-26 | 2015-05-20 | 江苏大学 | Fast tracking recognition method for overlapped fruits by picking robot |
| CN104658000A (en) * | 2015-03-08 | 2015-05-27 | 无锡桑尼安科技有限公司 | Orange yield estimation system based on multiple estimation parameters |
| CN104677342A (en) * | 2015-03-08 | 2015-06-03 | 无锡桑尼安科技有限公司 | Apple orchard yield measuring method |
| CN104881626A (en) * | 2015-01-19 | 2015-09-02 | 新疆农业大学 | Recognition method for fruit of fruit tree |
| CN106407962A (en) * | 2016-11-15 | 2017-02-15 | 融安县植保植检站 | Citrus fruit fly harm fruit-drop rate statistics system |
| CN108053078A (en) * | 2017-12-28 | 2018-05-18 | 深圳春沐源控股有限公司 | A kind of production prediction method, server and computer readable storage medium |
| CN108364235A (en) * | 2018-02-26 | 2018-08-03 | 深圳春沐源控股有限公司 | Method and system for determining fruit pre-picking plan and fruit picking system |
| CN108491846A (en) * | 2018-04-02 | 2018-09-04 | 湖南生物机电职业技术学院 | Ripening fruits machine identification method |
| CN108734054A (en) * | 2017-04-17 | 2018-11-02 | 湖南生物机电职业技术学院 | Unobstructed citrusfruit image-recognizing method |
| CN108960011A (en) * | 2017-05-23 | 2018-12-07 | 湖南生物机电职业技术学院 | The citrusfruit image-recognizing method of partial occlusion |
| CN109658379A (en) * | 2018-11-09 | 2019-04-19 | 广西慧云信息技术有限公司 | A method of orange yield is quickly calculated by picture |
| CN109937733A (en) * | 2019-03-28 | 2019-06-28 | 北京农业智能装备技术研究中心 | A kind of orchard yield automatic measurement mechanism |
| CN110060294A (en) * | 2019-04-30 | 2019-07-26 | 中国农业科学院农业环境与可持续发展研究所 | A kind of yield assessment method of fruit tree crop |
| CN110178481A (en) * | 2019-05-22 | 2019-08-30 | 青岛理工大学 | Bulbil identification and method of adjustment in a kind of great Jiang essence amount directional planting |
| CN111325767A (en) * | 2020-02-17 | 2020-06-23 | 杭州电子科技大学 | Method for synthesizing image set of citrus trees based on real scene |
| CN111369497A (en) * | 2020-02-18 | 2020-07-03 | 中国农业科学院农业资源与农业区划研究所 | Walking type tree fruit continuous counting method and device |
| CN111402318A (en) * | 2020-02-18 | 2020-07-10 | 中国农业科学院农业资源与农业区划研究所 | Method and device for rapidly estimating yield of fruits on tree |
| CN114910147A (en) * | 2021-12-14 | 2022-08-16 | 成都农业科技职业学院 | Internet of things-based maturity and yield estimation method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020102020A1 (en) * | 1999-02-01 | 2002-08-01 | Richard Qian | Method for image characterization using color and texture statistics with embedded spatial information |
| CN1731216A (en) * | 2005-08-19 | 2006-02-08 | 广州地理研究所 | A remote sensing detection and evaluation method for the area and production of large-area crop raising |
| CN102102988A (en) * | 2009-12-22 | 2011-06-22 | 中国农业科学院农业环境与可持续发展研究所 | Method, system and device for measuring crop yield information in real time |
-
2014
- 2014-08-05 CN CN201410381889.2A patent/CN104200193A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020102020A1 (en) * | 1999-02-01 | 2002-08-01 | Richard Qian | Method for image characterization using color and texture statistics with embedded spatial information |
| CN1731216A (en) * | 2005-08-19 | 2006-02-08 | 广州地理研究所 | A remote sensing detection and evaluation method for the area and production of large-area crop raising |
| CN102102988A (en) * | 2009-12-22 | 2011-06-22 | 中国农业科学院农业环境与可持续发展研究所 | Method, system and device for measuring crop yield information in real time |
Non-Patent Citations (2)
| Title |
|---|
| 钱建平 等，: ""基于双侧图像识别的单株苹果产量估测模型"", 《农业工程学报》 * |
| 钱建平: ""基于单株标识的数字化果园精准管理技术研究"", 《中国博士学位论文全文数据库，农业科技辑》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104881626A (en) * | 2015-01-19 | 2015-09-02 | 新疆农业大学 | Recognition method for fruit of fruit tree |
| CN104881626B (en) * | 2015-01-19 | 2017-12-29 | 新疆农业大学 | The recognition methods of Tree Fruit |
| CN104636722A (en) * | 2015-01-26 | 2015-05-20 | 江苏大学 | Fast tracking recognition method for overlapped fruits by picking robot |
| CN104636722B (en) * | 2015-01-26 | 2018-02-27 | 江苏大学 | A kind of overlapping fruit quick Tracking Recognition method of picking robot |
| CN104658000A (en) * | 2015-03-08 | 2015-05-27 | 无锡桑尼安科技有限公司 | Orange yield estimation system based on multiple estimation parameters |
| CN104677342A (en) * | 2015-03-08 | 2015-06-03 | 无锡桑尼安科技有限公司 | Apple orchard yield measuring method |
| CN104616004A (en) * | 2015-03-08 | 2015-05-13 | 无锡桑尼安科技有限公司 | Citrus yield estimation method based on multi-estimation parameters |
| CN106407962A (en) * | 2016-11-15 | 2017-02-15 | 融安县植保植检站 | Citrus fruit fly harm fruit-drop rate statistics system |
| CN108734054A (en) * | 2017-04-17 | 2018-11-02 | 湖南生物机电职业技术学院 | Unobstructed citrusfruit image-recognizing method |
| CN108960011A (en) * | 2017-05-23 | 2018-12-07 | 湖南生物机电职业技术学院 | The citrusfruit image-recognizing method of partial occlusion |
| CN108960011B (en) * | 2017-05-23 | 2021-12-03 | 湖南生物机电职业技术学院 | Partially-shielded citrus fruit image identification method |
| CN108053078A (en) * | 2017-12-28 | 2018-05-18 | 深圳春沐源控股有限公司 | A kind of production prediction method, server and computer readable storage medium |
| CN108364235A (en) * | 2018-02-26 | 2018-08-03 | 深圳春沐源控股有限公司 | Method and system for determining fruit pre-picking plan and fruit picking system |
| CN108491846A (en) * | 2018-04-02 | 2018-09-04 | 湖南生物机电职业技术学院 | Ripening fruits machine identification method |
| CN108491846B (en) * | 2018-04-02 | 2021-12-10 | 湖南生物机电职业技术学院 | Mature fruit machine identification method |
| CN109658379A (en) * | 2018-11-09 | 2019-04-19 | 广西慧云信息技术有限公司 | A method of orange yield is quickly calculated by picture |
| CN109937733A (en) * | 2019-03-28 | 2019-06-28 | 北京农业智能装备技术研究中心 | A kind of orchard yield automatic measurement mechanism |
| CN110060294A (en) * | 2019-04-30 | 2019-07-26 | 中国农业科学院农业环境与可持续发展研究所 | A kind of yield assessment method of fruit tree crop |
| CN110178481B (en) * | 2019-05-22 | 2020-08-21 | 青岛理工大学 | Bulbil recognition and adjustment method in precise directional planting of gingers |
| CN110178481A (en) * | 2019-05-22 | 2019-08-30 | 青岛理工大学 | Bulbil identification and method of adjustment in a kind of great Jiang essence amount directional planting |
| CN111325767A (en) * | 2020-02-17 | 2020-06-23 | 杭州电子科技大学 | Method for synthesizing image set of citrus trees based on real scene |
| CN111369497A (en) * | 2020-02-18 | 2020-07-03 | 中国农业科学院农业资源与农业区划研究所 | Walking type tree fruit continuous counting method and device |
| CN111402318A (en) * | 2020-02-18 | 2020-07-10 | 中国农业科学院农业资源与农业区划研究所 | Method and device for rapidly estimating yield of fruits on tree |
| CN111402318B (en) * | 2020-02-18 | 2023-04-07 | 中国农业科学院农业资源与农业区划研究所 | Method and device for rapidly estimating yield of fruits on tree |
| CN111369497B (en) * | 2020-02-18 | 2023-10-10 | 中国农业科学院农业资源与农业区划研究所 | Walking type tree fruit continuous counting method and device |
| CN114910147A (en) * | 2021-12-14 | 2022-08-16 | 成都农业科技职业学院 | Internet of things-based maturity and yield estimation method and device |
| CN114910147B (en) * | 2021-12-14 | 2023-10-24 | 成都农业科技职业学院 | Maturity and yield estimation method and device based on Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104200193A (en) | Fruit tree yield estimation method and device | |
| Petrellis | A smart phone image processing application for plant disease diagnosis | |
| US11521380B2 (en) | Shadow and cloud masking for remote sensing images in agriculture applications using a multilayer perceptron | |
| Liu et al. | A computer vision system for early stage grape yield estimation based on shoot detection | |
| CN110430443B (en) | Method and device for cutting video shot, computer equipment and storage medium | |
| US20230017425A1 (en) | System and method for determining damage on crops | |
| WO2017088365A1 (en) | Skin-colour detection method and apparatus | |
| Liu et al. | A vision-based robust grape berry counting algorithm for fast calibration-free bunch weight estimation in the field | |
| JP2017134585A (en) | Harvesting proper time determination support device and harvesting proper time determination support program | |
| CN114581801A (en) | Fruit tree identification and quantity monitoring method based on unmanned aerial vehicle data acquisition | |
| WO2007044037A1 (en) | Robust perceptual color identification | |
| CN107220647B (en) | Crop center point positioning method and system under blade crossing condition | |
| CN103093208A (en) | Method and system for fruit and vegetable recognition | |
| CN103778630A (en) | Field disease degree image identification method based on android | |
| Liu et al. | A lightweight method for grape berry counting based on automated 3D bunch reconstruction from a single image | |
| Díaz et al. | Grapevine buds detection and localization in 3D space based on structure from motion and 2D image classification | |
| Ye et al. | An image-based approach for automatic detecting tasseling stage of maize using spatio-temporal saliency | |
| Rahim et al. | Deep learning-based accurate grapevine inflorescence and flower quantification in unstructured vineyard images acquired using a mobile sensing platform | |
| CN104411229A (en) | Image processing device, image processing method, and image processing program | |
| CN110119675B (en) | Product identification method and device | |
| CN111767943A (en) | Mulching film identification method and device, electronic equipment and storage medium | |
| CN113313692B (en) | Automatic banana young plant identification and counting method based on aerial visible light image | |
| Setyawan et al. | Comparison of hsv and lab color spaces for hydroponic monitoring system | |
| CN113962258A (en) | Method, system and storage medium for identifying and preventing tobacco diseases | |
| CN106919740A (en) | Detect method, device and the electronic equipment of plant nitrogen content |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141210 |