CN106951905A - Apple identification and localization method on a kind of tree based on TOF camera - Google Patents
Apple identification and localization method on a kind of tree based on TOF camera Download PDFInfo
- Publication number
- CN106951905A CN106951905A CN201710082634.XA CN201710082634A CN106951905A CN 106951905 A CN106951905 A CN 106951905A CN 201710082634 A CN201710082634 A CN 201710082634A CN 106951905 A CN106951905 A CN 106951905A
- Authority
- CN
- China
- Prior art keywords
- image
- apple
- tof camera
- fruit
- identification
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 230000004807 localization Effects 0.000 title claims abstract description 21
- 235000013399 edible fruits Nutrition 0.000 claims abstract description 44
- 238000001914 filtration Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 7
- 239000003550 marker Substances 0.000 claims description 6
- 230000006872 improvement Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 235000011430 Malus pumila Nutrition 0.000 claims 12
- 235000015103 Malus silvestris Nutrition 0.000 claims 12
- 241000220225 Malus Species 0.000 claims 11
- 244000141359 Malus pumila Species 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000000605 extraction Methods 0.000 abstract description 4
- 230000000877 morphologic effect Effects 0.000 abstract description 3
- 238000003384 imaging method Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000003306 harvesting Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- 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
- G06T5/00—Image enhancement or restoration
-
- G06T5/70—
-
- 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
- G06T2207/10012—Stereo images
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- 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
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses apple identification and localization method on a kind of tree based on TOF camera.Its technical scheme is, the use of resolution ratio is 1280 (h) × 1024 (v), the 3 Dimension Image Technique of model OI VS 1000 TOF camera obtains the intensity of reflected light image and depth image of apple, and gained image is pre-processed, using selective contrast enhancement algorithms, depth image is improved, then Morphological Filtering Algorithm is used, filter the noise in depth map, the algorithm merged using regional split is split objective fruit region and background area, then Canny operator extractions border is utilized, loop truss finally is carried out to the Hough transform that shape is positioned in the imagery exploitation image of border, obtain fruit circle and home position, the three-dimensional coordinate in each fruit center of circle is extracted with reference to the location technology of TOF camera three-dimensional data base, try to achieve fruit real radius value, realize the identification and positioning to apple.
Description
Technical field
The present invention relates to apple identification and localization method on a kind of tree based on TOF camera, belong to image recognition technology neck
Domain.
Background technology
China is an apple cultivation big country, and cultivated area is wide, annual apple production rapid growth, therefore, intelligence
Change, information-based plant correlation detection technology arises at the historic moment, and the full segmentation of fruit, it is the basis of related algorithm,
It is one of core research topic of computer vision field.As market demand increases, apple establishing in large scale, in order to timely
The apple of harvesting ripe, one of the important channel solved the above problems is exactly the application of apple picking robot.
Machine vision is the maximum external environmental information source of picking robot, and its main task includes the knowledge of harvesting target
Not, positioning and the extraction of correlated characteristic.At present, technique of binocular stereoscopic vision, ultrasonic imaging and laser scanner technique are obtained
Concern and research.Technique of binocular stereoscopic vision realizes identification to fruit using color space, and by image processing techniques at
Overlapping and occlusion issue is managed, realizes and positions finally by Stereo matching.The region extracted in binocular image due to blocking and
The change of light causes same having differences property of target, causes depth information inaccurate, while Stereo Matching Algorithm is complicated, consumption
When.Ultrasonic imaging and laser scanner technique are realized by the analysis of three-dimensional feature and spectral absorption characteristics to fruit surface
Identification and positioning to fruit.Although this technological orientation high precision, poor real, high cost, it is impossible to generally use.
TOF camera is so that its image taking speed is fast, positioning precision is high, high-resolution the advantages of in environment sensing, navigation and target
The fields such as Tracking Recognition are widely applied, relative two dimensional image, can be by more being enriched between range information acquisition object
Position relationship, that is, distinguish prospect and background, the attitude to complex object judges extremely effective.The measurement of TOF camera is mainly sharp
With the physical propagation characteristic of light, infrared transmitter sends cosine-modulation signal to target scene, is reflected through object in scene
Received afterwards by photo-detector, the domain reflection-based amplitude of three-dimensional information at each pixel is calculated by " 4 mensurations ".
The content of the invention
The present invention is directed in above-mentioned existing vision system, the deficiency existed in terms of the discrimination of apple, it is proposed that one
Apple identification and localization method on kind discrimination height, the tree based on high-precision TOF camera of accurate positioning.
The present invention solves identification and uses following scheme with orientation problem：
The present invention is carried out as follows using apple identification on a kind of tree based on TOF camera and localization method：
Step 1：Using resolution ratio be 1280 (h) × 1024 (v), model OI-VS-1000 TOF camera it is three-dimensional into
As the intensity of reflected light image and depth image of the apple that technical limit spacing grows naturally, and image is pre-processed；
Step 2：The visual effect for the image that step 1 is obtained carries out enhancing improvement, suppresses garbage, then it is entered
Row filtering process.
Step 3：Intensity map and depth map to step 2 resulting improvement carry out split degree, take two parts common factor conduct
The target area of identification.Then rim detection is carried out to bianry image using Canny operators, extracts the profile of fruit image
Step 4：Obtained image is subjected to zone marker, then the minimum circumscribed circle of drawing area utilizes Hough transform
The contour line of single fruit is fitted, the identification to fruit is completed.
Step 5：The shape facilities such as central coordinate of circle, the radius of fruit can be extracted by step 4, it is three-dimensional using TOF camera
The characteristics of database, directly reads the three-dimensional coordinate in the center of circle, and then tries to achieve fruit real radius value, realizes the positioning to apple.
In step 1, obtain the depth map after system compensation by computer real-time acquisition processing routine and represent and reflect
The intensity map of light amplitude.The depth information of each test point to camera plane is converted to by each pixel of gray level image by formula (1)
Gray value, realizes the visualization of three-dimensional data.
In formula, g (i, j) represents the gray value of gray level image；Z (i, j) represents the depth value that sensor is obtained.
The type of pixel of gray level image is set as 16, preserving the three-dimensional of each pixel with the form of database sits
Mark, the application condition so done when can make to be converted into discrete gray value by continuous depth information is small.
During apple identification, the strength information combination depth information collected is corrected by formula (2), eliminated
The influence that object distance is produced to intensity map gray value.
A'(i, j)=A (i, j) z (i, j)2 (2)
In formula, A'(i is j) intensity level after correction；A (i, j) is raw intensity values.
In step 2 using selective contrast enhancement process.Scope ratio of the gray value of background in image histogram
It is larger, directly ignore corresponding element, the gray space [a, b] comprising important information is expanded, passes through formula (3), i.e. tonal range
Just completed from [a, b] to the gradation conversion of [0,255].Then shape filtering processing is carried out again removes noise.
In formula, G (x, y) is gray value, and G'(x is y) gray value after treatment.
In step 3, the intensity map and depth map obtained to step 2 carries out splitting and merging respectively.Comprise the concrete steps that,
R is made to represent whole intensity image region, P represents certain similarity criterion.Image is divided into 4 regions first, then repeatedly
The subgraph that segmentation is obtained is again split into 4 regions, until to any Ri, P (Ri)=TRUE, represents region RiMeet
Similarity criterion, now no longer carries out splitting operation.If P (Ri)=FALSE, then by RiIt is divided into 4 regions.So continue
Go down, until P (Ri)=TRUE has arrived single pixel., will be any in image after releasing or while division
Two adjacent area R with similar featuresj、RkMerge, if i.e. P (Rj∪Rk)=TRUE, then merge Rj、Rk.When can not be again
Operate and stop when being polymerize or being divided.Then rim detection is carried out to bianry image using Canny operators, extracts fruit image
Profile.
In step 4, zone marker, and the minimum circumscribed circle of drawing area are carried out to image first, is pressed in circumscribed circle
Carry out circular Hough transform according to the parameter area of determination the profile of single fruit is fitted, then remove and recognized fruit
Contour line, is fitted to remaining Apple profile, all fruits is identified successively.
In steps of 5, fruit circle and home position can be obtained by carrying out loop truss using Hough transform by step 4, be utilized
The algorithm of TOF camera three-dimensional data base is extracted to the three-dimensional coordinate in the fruit center of circle, then tries to achieve fruit real radius value, real
Now apple is accurately positioned.
The image shot with other cameras is compared, and advantage of the invention is embodied in：
1st, the present invention proposes to use resolution ratio for 1280 (h) × 1024 (v), model OI-VS-1000 TOF camera, this
Model TOF camera has that image taking speed is fast, positioning precision is high, high-resolution the characteristics of.Relative two dimensional image, can be believed by distance
Breath obtains the position relationship more enriched between object, that is, distinguishes prospect and background, and the attitude to complex object judges extremely have
Effect.The measurement of TOF camera mainly make use of the physical propagation characteristic of light, and infrared transmitter sends cosine-modulation to target scene
Signal, through being received in scene after object reflection by photo-detector, the three-dimensional at each pixel is calculated by " 4 mensurations "
Information and reflected light amplitude.
2nd, the present invention uses selective contrast enhancement algorithms and Morphological Filtering Algorithm respectively in terms of enhancing and denoising, and
Image is improved well；And the algorithm of split degree is used intensity map and depth map respectively, two parts are occured simultaneously
As the target area of identification, using Canny operator extraction objective contours, according to the features of shape of fruit, circular Hough is utilized
Conversion identification fruit.
3rd, the present invention can not only realize the identification to fruit using the high-resolution TOF camera of OI-VS-1000 models,
And in positioning,, can be quickly and accurately to identification in conjunction with the information of three-dimensional data base according to the step of above-mentioned identification apple
The apple gone out is positioned.
Brief description of the drawings
Fig. 1 is the flow chart of apple identification and localization method on a kind of tree based on TOF camera of the invention provided；
Fig. 2 is that apple identification and localization method pass through on a kind of tree based on TOF camera provided in an embodiment of the present invention
TOF camera and the intensity map (a) of treated acquisition and depth map (b)；
Fig. 3 is the contrast of apple identification and localization method on a kind of tree based on TOF camera provided in an embodiment of the present invention
Spend enhanced depth image；
Fig. 4 is the division of apple identification and localization method on a kind of tree based on TOF camera provided in an embodiment of the present invention
Carry out the figure of rim detection after merging to image through Canny algorithms；
Fig. 5 be on a kind of tree based on TOF camera provided in an embodiment of the present invention apple identification and localization method through circle
The Apple identified after shape Hough transform；
Fig. 6 is the fruit of apple identification and localization method on a kind of tree based on TOF camera provided in an embodiment of the present invention
DLE schematic diagram.
Embodiment
The method of the present invention is further described below in conjunction with accompanying drawing, the present embodiment is premised on technical solution of the present invention
It is lower to be implemented, give detailed embodiment.It is emphasized that what the description below was merely exemplary, rather than be
Limitation the scope of the present invention and application.
As shown in figure 1, the embodiment of the present invention illustrates a kind of based on TOF phases exemplified by based on the image that TOF camera is gathered
Apple identification and localization method on the tree of machine.Detailed process is as follows：
Step 1：Depth map after system compensation is obtained by computer real-time acquisition processing routine and reflected light width is represented
The intensity map of value, and image is pre-processed.The depth information of each test point to camera plane is converted to by ash by formula (1)
The gray value of each pixel of image is spent, the visualization of three-dimensional data is realized.
In formula, g (i, j) represents the gray value of gray level image；Z (i, j) represents the depth value that sensor is obtained.
The type of pixel of gray level image is set as 16, preserving the three-dimensional of each pixel with the form of database sits
Mark, the application condition so done when can make to be converted into discrete gray value by continuous depth information is small.
During apple identification, the strength information combination depth information collected is corrected by formula (2), eliminated
The influence that object distance is produced to intensity map gray value.
A'(i, j)=A (i, j) z (i, j)2 (2)
In formula, A'(i is j) intensity level after correction；A (i, j) is raw intensity values, and apple is obtained according to the present embodiment
The correction intensity map and depth map of fruit are as shown in Figure 2.
Step 2：The image obtained to step 1 carries out selective contrast enhancement processing, improves image, suppresses unrelated letter
Breath, then carries out shape filtering processing to the image after improvement, removes the noise in image.
Using selective contrast enhancement process.Scope of the gray value of background in image histogram is than larger, directly
Ignore corresponding element, the gray space [a, b] comprising important information is expanded, arrived by formula (3), i.e. tonal range from [a, b]
[0,255] gradation conversion is completed.
In formula, G (x, y) is gray value, and G'(x is y) gray value after treatment.
By formula (4), shape filtering processing is carried out to image, structural element B is 3 × 3 square templates, filter factor n
=3, i.e., first corrode 3 times, reflation 3 times.Image F passes through filtering, you can eliminate noise, image G after being improved, according to this reality
The Apple image for applying example acquisition improvement is as shown in Figure 3.
Step 3：The image improved to step 2 carries out split degree, the area obtained when carrying out split degree to intensity map
Domain can be by the region merging technique of adjacent apple to same region, and obtained region can be by when carrying out split degree by depth map
The leaf for being attached at apple surface is merged into same region, therefore, the mesh for taking the common factor that two parts merge to be obtained as identification
Region is marked, rim detection is then carried out with Canny operators, the profile in objective fruit region is obtained.
The intensity map and depth map obtained to step 2 carries out splitting and merging respectively.R is made to represent whole intensity image area
Domain, P represents certain similarity criterion.Image is divided into 4 regions first, the subgraph for then repeatedly obtaining segmentation is again
It is divided into 4 regions, until to any Ri, P (Ri)=TRUE, represents region RiSimilarity criterion has been met, has now no longer been carried out
Splitting operation.If P (Ri)=FALSE, then by RiIt is divided into 4 regions.So continue, until P (Ri)=TRUE or
Single pixel is arrived.After releasing or while division, any two in image is had to the adjacent region of similar features
Domain Rj、RkMerge, if i.e. P (Rj∪Rk)=TRUE, then merge Rj、Rk.When can not be polymerize again or time-sharing operation stop.So
Rim detection is carried out to bianry image using Canny operators afterwards, the profile of fruit image is extracted, point obtained according to the present embodiment
Split merging figure as shown in Figure 4.
Step 4：Zone marker, and the minimum circumscribed circle of drawing area are carried out to the image obtained by step 3, external
The contour line of single fruit is fitted by circular Hough transform in circle, all fruits are identified successively.
Zone marker, and the minimum circumscribed circle of drawing area are carried out to image first, according to the ginseng of determination in circumscribed circle
Number scope carries out circular Hough transform and the profile of single fruit is fitted, then removes the contour line for having recognized fruit, to surplus
Remaining Apple profile is fitted, and all fruits are identified successively, the apple figure such as Fig. 5 institutes identified according to the present embodiment
Show.
Step 5：The home position of apple can be obtained by carrying out loop truss using Hough transform according to step 4, further according to
The three-dimensional imaging location technology algorithm of TOF camera obtains the three-dimensional coordinate of apple, asks for the real radius value of apple, realizes to apple
The positioning of fruit, fruit DLE schematic diagram is as shown in Figure 6.The image-forming principle and pinhole imaging system of TOF camera first are somewhat like,
According to similar triangle theory, the radius formula of apple can be obtained, as shown in formula (5)：
In formula, RcmTo calculate obtained apple radius；DiFor TOF camera pel spacing；RpxFor the pixel of minimum circumscribed circle
Radius；F is TOF camera focal length；Z is object distance.
By looking into the parameter of TOF camera, convolution (1) can obtain radius value and depth image gray value and pixel radius
Between relational expression, as shown in Equation 6：
Obtained after the central coordinate of circle of apple, obtained often further according to calculating by the distinctive three-dimensional imaging location technology of TOF camera
Individual apple radius value, just can realize being accurately positioned for each apple.
Based on the present embodiment uses resolution ratio for 1280 (h) × 1024 (v), model OI-VS-1000 TOF camera,
Realize apple identification and localization method on the tree based on TOF camera.This method can not only be recognized and oriented under simple scenario
Apple situation, and be still applicable with overlapping situation having to block.The present invention is favorably improved the discrimination of fruit, improves
The deficiency existed all the time in machine vision this respect before, identification and positioning for fruit from now on provide good
Method.
To sum up, apple identification and localization method on a kind of tree based on TOF camera of the invention.The use of resolution ratio is 1280
(h) × 1024 (v), model OI-VS-1000 TOF camera 3 Dimension Image Technique obtain apple intensity of reflected light image
And depth image, and gained image is pre-processed, using selective contrast enhancement algorithms, depth image is changed
It is kind, then using Morphological Filtering Algorithm, the noise in depth map is filtered, the algorithm merged using regional split is by objective fruit area
Domain is split with background area, then using Canny operator extractions border, finally to positioning shape in the imagery exploitation image of border
Hough transform carry out loop truss, obtain fruit circle and home position, with reference to TOF camera three-dimensional data base location technology to every
The three-dimensional coordinate in the individual fruit center of circle is extracted, and tries to achieve fruit real radius value, realizes identification and positioning to apple.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " illustrative examples ",
The description of " example ", " specific example " or " some examples " etc. means to combine specific features, the knot that the embodiment or example are described
Structure, material or feature are contained at least one embodiment of the present invention or example.In this manual, to above-mentioned term
Schematic representation is not necessarily referring to identical embodiment or example.Moreover, specific features, structure, material or the spy of description
Point can in an appropriate manner be combined in any one or more embodiments or example.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：Not
In the case of departing from the principle and objective of the present invention a variety of change, modification, replacement and modification can be carried out to these embodiments, this
The scope of invention is limited by claim and its equivalent.
Claims (6)
1. apple identification and localization method on a kind of tree based on TOF camera, it is characterised in that：Comprise the following steps：
Step 1：The intensity of reflected light image and depth map of the apple tree that nature grows are obtained using the 3 Dimension Image Technique of camera
Picture, and image is pre-processed；Step 2：The image obtained to step 1 carries out selective contrast enhancement processing, improves figure
Picture, suppresses irrelevant information, then carries out shape filtering processing to the image after improvement, removes the noise in image；Step 3：It is right
The image that step 2 improved carries out split degree, and the region obtained when carrying out split degree to intensity map can be by adjacent apple
Region merging technique is to same region, and the region obtained when carrying out split degree by depth map can will be attached at apple surface
Leaf is merged into same region, therefore, the target area for taking the common factor that two parts merge to be obtained as identification, Ran Houyong
Canny operators carry out rim detection, obtain the profile in objective fruit region；Step 4：The image obtained by step 3 is carried out
Zone marker, and the minimum circumscribed circle of drawing area, pass through contour line of the circular Hough transform to single fruit in circumscribed circle
It is fitted, all fruits is identified successively；Step 5：Fruit circle and home position can be obtained by step 4, then basis
The three-dimensional data base of camera is extracted to the three-dimensional coordinate in each fruit center of circle, then seeks the real radius value of apple, is realized
Positioning to apple.
2. apple identification and localization method on a kind of tree based on TOF camera according to claim 1, it is characterised in that
In the step 1 obtain image method be, computer by timely collection program obtain system compensation after depth map and
The intensity map of reflected light amplitude, is then converted to the gray-scale map of each pixel of gray level image.
3. apple identification and localization method on a kind of tree based on TOF camera according to claim 1, it is characterised in that
It is because gray value is background information in a big way, these information are straight using selective contrast enhancement process in the step 2
Connect and ignore, gray space of the extension packets containing important information can thus be transformed into full gray space；Skill is handled using shape filtering
Art, i.e., first corrode 3 times, reflation 3 times, so that it may eliminate noise.
4. apple identification and localization method on a kind of tree based on TOF camera according to claim 1, it is characterised in that
The specific method of split degree is to make R represent whole intensity image region in the step 3, and P represents certain similarity criterion,
Image is divided into 4 regions first, the subgraph for then repeatedly obtaining segmentation is again split into 4 regions, until to any
Ri, P (Ri)=TRUE, represents region RiSimilarity criterion has been met, splitting operation is now no longer carried out；If P (Ri)=
FALSE, then by RiIt is divided into 4 regions；So continue, until P (Ri)=TRUE has arrived single pixel；Dividing
From after or while division, any two in image is had to the adjacent area R of similar featuresj、RkMerge, if i.e. P
(Rj∪Rk)=TRUE, then merge Rj、Rk；Operate and stop when that can not be polymerize or be divided again.
5. apple identification and localization method on a kind of tree based on TOF camera according to claim 1, it is characterised in that
In the step 4, zone marker, and the minimum circumscribed circle of drawing area are carried out first, according to the parameter of determination in circumscribed circle
Scope carries out circular Hough transform and the contour shape of single fruit is fitted, then removes the contour line for having recognized fruit, right
Remaining Apple contour line is fitted, and all fruits are identified successively.
6. apple identification and localization method on a kind of tree based on TOF camera according to claim 1, it is characterised in that
The resolution ratio of the TOF camera is 1280 (h) × 1024 (v), model OI-VS-1000 TOF camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710082634.XA CN106951905A (en) | 2017-02-16 | 2017-02-16 | Apple identification and localization method on a kind of tree based on TOF camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710082634.XA CN106951905A (en) | 2017-02-16 | 2017-02-16 | Apple identification and localization method on a kind of tree based on TOF camera |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106951905A true CN106951905A (en) | 2017-07-14 |
Family
ID=59467661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710082634.XA Pending CN106951905A (en) | 2017-02-16 | 2017-02-16 | Apple identification and localization method on a kind of tree based on TOF camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106951905A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107984482A (en) * | 2017-12-27 | 2018-05-04 | 中国科学院合肥物质科学研究院 | A kind of intelligent matrimony vine picker and method based on TOF depth perceptions |
| CN109871900A (en) * | 2019-03-06 | 2019-06-11 | 哈尔滨理工大学 | The recognition positioning method of apple under a kind of complex background based on image procossing |
| CN110717408A (en) * | 2019-09-20 | 2020-01-21 | 台州智必安科技有限责任公司 | People flow counting method based on TOF camera |
| CN110853080A (en) * | 2019-09-30 | 2020-02-28 | 广西慧云信息技术有限公司 | Method for measuring size of field fruit |
| CN111126314A (en) * | 2019-12-26 | 2020-05-08 | 杭州中科先进技术研究院有限公司 | Passenger flow statistical visual method and system based on 3D point cloud data |
| CN111507339A (en) * | 2020-04-16 | 2020-08-07 | 北京深测科技有限公司 | Target point cloud obtaining method based on intensity image |
| CN111738146A (en) * | 2020-06-22 | 2020-10-02 | 哈尔滨理工大学 | Rapid separation and identification method for overlapped fruits |
| WO2021042693A1 (en) * | 2019-09-04 | 2021-03-11 | 五邑大学 | Mining process-based method for acquiring three-dimensional coordinates of ore and apparatus therefor |
| CN112927154A (en) * | 2021-03-05 | 2021-06-08 | 上海炬佑智能科技有限公司 | ToF device, depth camera and gray scale image enhancement method |
| CN113466320A (en) * | 2021-08-06 | 2021-10-01 | 河北省食品检验研究院 | MALDI-TOF detection method for bovine-derived components in milk powder |
| CN113658089A (en) * | 2021-09-09 | 2021-11-16 | 南开大学 | Double-data-stream fusion object identification method based on depth camera |
| CN114067206A (en) * | 2021-11-16 | 2022-02-18 | 哈尔滨理工大学 | Spherical fruit identification and positioning method based on depth image |
| CN114260895A (en) * | 2021-12-22 | 2022-04-01 | 江苏大学 | Method and system for determining grabbing obstacle avoidance direction of mechanical arm of picking machine |
| CN114067206B (en) * | 2021-11-16 | 2024-04-26 | 哈尔滨理工大学 | Spherical fruit identification positioning method based on depth image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101726251A (en) * | 2009-11-13 | 2010-06-09 | 江苏大学 | Automatic fruit identification method of apple picking robot on basis of support vector machine |
| CN102831398A (en) * | 2012-07-24 | 2012-12-19 | 中国农业大学 | Tree apple recognition method based on depth image |
-
2017
- 2017-02-16 CN CN201710082634.XA patent/CN106951905A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101726251A (en) * | 2009-11-13 | 2010-06-09 | 江苏大学 | Automatic fruit identification method of apple picking robot on basis of support vector machine |
| CN102831398A (en) * | 2012-07-24 | 2012-12-19 | 中国农业大学 | Tree apple recognition method based on depth image |
Non-Patent Citations (2)
| Title |
|---|
| 孙力等: "基于TOF成像技术的柑橘实时识别与定位", 《江苏大学学报(自然科学版)》 * |
| 项荣: "开放环境中番茄的双目立体视觉识别与定位", 《中国博士学位论文全文数据库 农业科技辑》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107984482A (en) * | 2017-12-27 | 2018-05-04 | 中国科学院合肥物质科学研究院 | A kind of intelligent matrimony vine picker and method based on TOF depth perceptions |
| CN109871900A (en) * | 2019-03-06 | 2019-06-11 | 哈尔滨理工大学 | The recognition positioning method of apple under a kind of complex background based on image procossing |
| WO2021042693A1 (en) * | 2019-09-04 | 2021-03-11 | 五邑大学 | Mining process-based method for acquiring three-dimensional coordinates of ore and apparatus therefor |
| CN110717408B (en) * | 2019-09-20 | 2022-04-12 | 台州智必安科技有限责任公司 | People flow counting method based on TOF camera |
| CN110717408A (en) * | 2019-09-20 | 2020-01-21 | 台州智必安科技有限责任公司 | People flow counting method based on TOF camera |
| CN110853080A (en) * | 2019-09-30 | 2020-02-28 | 广西慧云信息技术有限公司 | Method for measuring size of field fruit |
| CN111126314A (en) * | 2019-12-26 | 2020-05-08 | 杭州中科先进技术研究院有限公司 | Passenger flow statistical visual method and system based on 3D point cloud data |
| CN111507339A (en) * | 2020-04-16 | 2020-08-07 | 北京深测科技有限公司 | Target point cloud obtaining method based on intensity image |
| CN111507339B (en) * | 2020-04-16 | 2023-07-18 | 北京深测科技有限公司 | Target point cloud acquisition method based on intensity image |
| CN111738146A (en) * | 2020-06-22 | 2020-10-02 | 哈尔滨理工大学 | Rapid separation and identification method for overlapped fruits |
| CN112927154A (en) * | 2021-03-05 | 2021-06-08 | 上海炬佑智能科技有限公司 | ToF device, depth camera and gray scale image enhancement method |
| CN112927154B (en) * | 2021-03-05 | 2023-06-02 | 上海炬佑智能科技有限公司 | ToF device, depth camera and gray image enhancement method |
| CN113466320B (en) * | 2021-08-06 | 2022-10-14 | 河北省食品检验研究院 | MALDI-TOF detection method for bovine-derived components in milk powder |
| CN113466320A (en) * | 2021-08-06 | 2021-10-01 | 河北省食品检验研究院 | MALDI-TOF detection method for bovine-derived components in milk powder |
| CN113658089A (en) * | 2021-09-09 | 2021-11-16 | 南开大学 | Double-data-stream fusion object identification method based on depth camera |
| CN114067206A (en) * | 2021-11-16 | 2022-02-18 | 哈尔滨理工大学 | Spherical fruit identification and positioning method based on depth image |
| CN114067206B (en) * | 2021-11-16 | 2024-04-26 | 哈尔滨理工大学 | Spherical fruit identification positioning method based on depth image |
| CN114260895A (en) * | 2021-12-22 | 2022-04-01 | 江苏大学 | Method and system for determining grabbing obstacle avoidance direction of mechanical arm of picking machine |
| CN114260895B (en) * | 2021-12-22 | 2023-08-22 | 江苏大学 | Method and system for determining grabbing obstacle avoidance direction of mechanical arm of picking robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106951905A (en) | Apple identification and localization method on a kind of tree based on TOF camera | |
| CN110569704B (en) | Multi-strategy self-adaptive lane line detection method based on stereoscopic vision | |
| CN102704215B (en) | Automatic cutting method of embroidery cloth based on combination of DST file parsing and machine vision | |
| CN105260699B (en) | A kind of processing method and processing device of lane line data | |
| CN109146948B (en) | Crop growth phenotype parameter quantification and yield correlation analysis method based on vision | |
| CN106296666B (en) | A kind of color image removes shadow method and application | |
| CN111340797A (en) | Laser radar and binocular camera data fusion detection method and system | |
| CN110647850A (en) | Automatic lane deviation measuring method based on inverse perspective principle | |
| CN109460709A (en) | The method of RTG dysopia analyte detection based on the fusion of RGB and D information | |
| CN107248159A (en) | A kind of metal works defect inspection method based on binocular vision | |
| CN107392929B (en) | Intelligent target detection and size measurement method based on human eye vision model | |
| CN104700404A (en) | Fruit location identification method | |
| CN108197583A (en) | The building change detecting method of optimization and image structure feature is cut based on figure | |
| CN103870834B (en) | Method for searching for sliding window based on layered segmentation | |
| CN107230203A (en) | Casting defect recognition methods based on human eye vision attention mechanism | |
| CN108960011B (en) | Partially-shielded citrus fruit image identification method | |
| CN110232379A (en) | A kind of vehicle attitude detection method and system | |
| CN108596165A (en) | Road traffic marking detection method based on unmanned plane low latitude Aerial Images and system | |
| CN105913013A (en) | Binocular vision face recognition algorithm | |
| CN104268853A (en) | Infrared image and visible image registering method | |
| CN110070571B (en) | Phyllostachys pubescens morphological parameter detection method based on depth camera | |
| CN109559324A (en) | A kind of objective contour detection method in linear array images | |
| CN106803262A (en) | The method that car speed is independently resolved using binocular vision | |
| Tiede et al. | Process oriented object-based algorithms for single tree detection using laser scanning | |
| CN114140665A (en) | Dense small target detection method based on improved YOLOv5 |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170714 |
|
| RJ01 | Rejection of invention patent application after publication |