CN104751455A - Crop image dense matching method and system - Google Patents
Crop image dense matching method and system Download PDFInfo
- Publication number
- CN104751455A CN104751455A CN201510111324.7A CN201510111324A CN104751455A CN 104751455 A CN104751455 A CN 104751455A CN 201510111324 A CN201510111324 A CN 201510111324A CN 104751455 A CN104751455 A CN 104751455A
- Authority
- CN
- China
- Prior art keywords
- dense matching
- crop
- matching
- target
- dense
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a crop image dense matching method which includes the following steps: calculating a matching point of a pixel in a central area of a target to be matched in a left image in a right image; intercepting sub images of two images with the target center point and the matching point as the center based on the target center matching point; eliminating images on the outside of an analysis target boundary through a boundary partitioning algorithm in the two sub images; for pixel points of each analysis target, searching for pixel points closest to the STFT vector in the right sub image to serve as the matching points to further obtain the matching point of each pixel point and obtain dense matching of a target crop; acquiring a source image dense matching result through inverse transformation of the acquired sub images based on the dense matching result of the sub images. By means of the matching method and system, the dense matching result can be accurately achieved for shot crop images, the aim of crop state analysis is achieved, and the method and system provides a basis for further real-time analysis and treatment.
Description
Technical field
The present invention relates to Agricultural information field, particularly a kind of crop map is as dense matching method and system.
Background technology
Images match is the important component part of pattern-recognition, computer vision field.The exact matching of image is the basis of the subsequent treatment work such as image object location, motion detection, three-dimensional reconstruction, information fusion, and the quality of its result will directly affect the quality of subsequent analysis work.
And in crop graphical analysis, in order to crop condition information can be analyzed, as the information such as shape, leaf blade size, fruit size, plant height of ill blade damage area, need to carry out dense matching calculating to crop map picture, namely obtain the position corresponding relation of each pixel in multiple image of target image.But although at industrial circle, dense matching research has had certain achievement in research, in agricultural crops image, due to the similarity of the color of crop map picture and background, brightness and texture, the dense matching of crop map picture is a complicated job.In the dense matching for crop map picture, be in progress less.
At present, the research of dense matching mainly contains based on the method for optical flow field and the matching process based on SIFT stream.
So-called optical flow field, is when having relative motion when between video camera and scene objects, gives a velocity to each pixel in image, obtain a sports ground about image and be called optical flow field.In optical flow field research, up to the present, various method and improve one's methods nearly tens kinds, is conceptually divided into roughly 4 classes: based on gradient method, based on region method, based on ENERGY METHOD with based on phase method.Although achieve larger achievement based on the dense matching algorithm of optical flow field in the application aspect of industry, but the crop map gathered under physical environment is often more complicated than industrial dense matching as dense matching, the color, the brightness that mainly there is different leaves, different fruit and background are similar, different shooting angles also exists the reason such as difference of blocking difference, different angles crop reflected light photograph, optical flow field based on brightness uniformity in crop map picture, be difficult to application, thus had a strong impact on the precision of optical flow computation.
SIFT matching algorithm is a kind of matching algorithm based on metric space that David proposed in 2004.SIFT has stable matching capacity under image scaling, rotation, brightness change, is acknowledged as one of the most successful two matching algorithms during the nearly last ten years, and another matching algorithm is the surf matching algorithm that Bay proposed in 2008.But SIFT matching algorithm and surf matching algorithm all only find the most matching double points in two width images, are not dense matching algorithms.CeLiu etc. by means of the thought of optical flow field, and the SIFT proposed based on SIFT feature vector consistency flows dense matching method.The basis of SIFT stream calculation is the shade of gray consistance of architectural feature.In the simple industrial application of some structures, SIFT stream can reach higher matching effect accurately.But in agriculture field, under physical environment, the structure of crop map picture is very complicated, crop each several part and background similarity high, thus have impact on the SIFT matching effect of crop map picture.Have rotational invariance because SIFT calculates, fringe region and the non-matching dot structure of crop map picture also may have larger similarity.Inventor calculates the dense matching of sweet potato leaves, strawberry and eggplant leaf image with several typical optical flow field algorithm and SIFT flow algorithm, and experimental result display exists larger gap, is difficult to meet crop analysis needs.
Be at present almost blank based on crop map complicated under physical environment as the research of dense matching, this is mainly because the complex structure of crop map picture under physical environment, the reason such as in disorder that distributes cause.The most industrial application of some important achievements at present, is difficult to application in agriculture field, cannot reach the object that crop condition is analyzed.
Summary of the invention
The object of the invention is to overcome the shortcoming of prior art and deficiency, provide a kind of crop map as dense matching method.
Another object of the present invention is to provide a kind of crop map as dense matching system.
Object of the present invention is realized by following technical scheme:
A kind of crop map, as dense matching method, comprises the step of following order:
S1. the center coupling of evaluating objects: to a pixel of the central area of left figure target to be matched, go out its match point in right figure by SIFT Flow Field Calculation;
S2. evaluating objects dividing sub-picture: on the basis obtaining target's center's match point, intercept the subgraph of two width images centered by target's center's point and its match point;
S3. subgraph boundary segmentation and subgraph dense matching: in two subgraphs, removes the image outside evaluating objects border by boundary segmentation algorithm; To the pixel of each evaluating objects, in right subgraph, search and the pixel of its SIFT vector minimum distance are as match point, thus obtain each pixel matching point, obtain the dense matching of target crop;
S4. the dense matching in former figure: on the basis of subgraph dense matching result, the inverse transformation of the subgraph intercepted by step S2, obtains source images dense matching result.
In step S2, the size of described target dividing sub-picture will set according to the size of monitoring objective.To ensure that the target of Water demand can both completely occur in two width subgraphs.
In step S3, described boundary segmentation algorithm is canny boundary segmentation algorithm.
Another object of the present invention is realized by following technical scheme:
A kind of crop map, as dense matching system, comprises thin client, the camera be connected with thin client, an AP, also comprises the supply module for thin client, an AP power, and server, the 2nd AP that is connected with server.
Described supply module comprises controller, and the accumulator, solar panels, the inverter that are connected with controller respectively, and described inverter is connected with thin client, an AP respectively.Crop map of the present invention is in as the thin client of dense matching system, camera, an AP by the crop that will monitor, is generally field, and therefore supply module adopts sun power can be more convenient, need not lay power cable specially, also relative energy-saving environmental protection.
Described camera is two Logitch Pro9000 cameras, and two camera distance 6 ~ 12 centimetres, ground, two camera directions are consistent.Prove with more test, Logitch Pro9000 camera can gather the crop map picture of better quality under different illumination conditions.Adjustment left and right cameras position and shooting direction.Two video cameras greatly about sustained height, distance 6-12 centimetre; Suitable adjustment camera direction, reaches in two video cameras and can both obtain the more complete image of target, then fixes position and the shooting direction of two video cameras.
Compared with prior art, tool has the following advantages and beneficial effect in the present invention:
1, the present invention mainly studies the dense matching method of crop map picture under physical environment.Achievement in research most is at present mainly in industrial application, and these achievements cannot directly apply to crop analysis object.The invention is under physical environment, and the crop map picture for shooting can reach more accurately dense matching result, can meet the object that crop condition is analyzed, for further real-time analysis and process provide basis.
2, the analysis of crop map picture to crop condition has extremely important effect, can obtain the growth conditions of crop and the information of disease and pest by crop graphical analysis.Along with the development of embedded technology and Radio Transmission Technology, people can gather the image of crop whenever and wherever possible and transmit, thus can obtain the image information of a large amount of crops.But abundant information and knowledge is deficient, the crop nanoscale regime information that the shape that people wish to obtain disease and pest region from a large amount of image gathered is enriched with size, blade and fruit size and plant height, crop dynamic changing process etc.But the image dense matching algorithm due to current existence cannot meet the accuracy requirement that crop condition is analyzed, and people also cannot obtain abundant crop dimensional information from image.
Inventor is found by a large amount of crop imaging experiments, affect the principal element of crop map as dense matching to cause due to the border such as blade, fruit, the invention, from this feature, by the subgraph dense matching method of evaluating objects, reaches the object of crop dense matching.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of a kind of crop map of the present invention as dense matching method;
Fig. 2 is the structural representation of a kind of crop map of the present invention as dense matching system.
Embodiment
Below in conjunction with embodiment and accompanying drawing, the present invention is described in further detail, but embodiments of the present invention are not limited thereto.
As Fig. 1, a kind of crop map, as dense matching method, comprises the step of following order:
S1. the center coupling of evaluating objects: to a pixel of the central area of left figure target to be matched, go out its match point in right figure by SIFT Flow Field Calculation;
S2. evaluating objects dividing sub-picture: on the basis obtaining target's center's match point, intercept the subgraph of two width images centered by target's center's point and its match point;
S3. subgraph boundary segmentation and subgraph dense matching: in two subgraphs, removes the image outside evaluating objects border by boundary segmentation algorithm; To the pixel of each evaluating objects, in right subgraph, search and the pixel of its SIFT vector minimum distance are as match point, thus obtain each pixel matching point, obtain the dense matching of target crop;
S4. the dense matching in former figure: on the basis of subgraph dense matching result, the inverse transformation of the subgraph intercepted by step S2, obtains source images dense matching result.
In step S2, the size of described target dividing sub-picture will set according to the size of monitoring objective.
In step S3, described boundary segmentation algorithm is canny boundary segmentation algorithm.
As Fig. 2, a kind of crop map, as dense matching system, comprises thin client, the camera be connected with thin client, an AP, also comprises the supply module for thin client, an AP power, and server, the 2nd AP that is connected with server.
Described supply module comprises controller, and the accumulator, solar panels, the inverter that are connected with controller respectively, and described inverter is connected with thin client, an AP respectively.
Described camera is two Logitch Pro9000 cameras, and two camera distance 6 ~ 12 centimetres, ground, two camera directions are consistent.
Specific as follows:
Utilize solar electric power supply system to power for wireless transmission and image capturing system, solar electric power supply system turns 220V500W pure sine wave inverter, 12V100AH lead-acid accumulator and controller for solar by 18V100W mono-crystalline silicon solar plate, 12V and forms.Utilize an Ap, the remote point-to-point Internet of the 2nd Ap connects and realize wireless transmission.
The equipment gathering crop map picture in embodiment is two Logitch Pro9000 cameras, proves with more test, and Logitch Pro9000 camera can gather the crop map picture of better quality under different illumination conditions.Adjustment left and right cameras position and shooting direction.Two video cameras greatly about sustained height, distance 6-12 centimetre.The direction of two video cameras is roughly consistent.Suitable adjustment camera direction, reaches in two video cameras and can both obtain the more complete image of target, then fixes position and the shooting direction of two video cameras.
Two Logitch Pro9000 camera USB interface and Tian Yuan risen and create board thin client and be connected simultaneously, thin client installs Window operating system.The collection of crop map picture requires that thin client controls two cameras and captures crop map picture simultaneously.Capture software VC to call OpenCV and encoded simultaneously.
Above-described embodiment is the present invention's preferably embodiment; but embodiments of the present invention are not restricted to the described embodiments; change, the modification done under other any does not deviate from Spirit Essence of the present invention and principle, substitute, combine, simplify; all should be the substitute mode of equivalence, be included within protection scope of the present invention.
Claims (6)
1. crop map is as a dense matching method, it is characterized in that, comprises the step of following order:
S1. the center coupling of evaluating objects: to a pixel of the central area of left figure target to be matched, go out its match point in right figure by SIFT Flow Field Calculation;
S2. evaluating objects dividing sub-picture: on the basis obtaining target's center's match point, intercept the subgraph of two width images centered by target's center's point and its match point;
S3. subgraph boundary segmentation and subgraph dense matching: in two subgraphs, removes the image outside evaluating objects border by boundary segmentation algorithm; To the pixel of each evaluating objects, in right subgraph, search and the pixel of its SIFT vector minimum distance are as match point, thus obtain each pixel matching point, obtain the dense matching of target crop;
S4. the dense matching in former figure: on the basis of subgraph dense matching result, the inverse transformation of the subgraph intercepted by step S2, obtains source images dense matching result.
2. crop map according to claim 1 is as dense matching method, it is characterized in that, in step S2, the size of described target dividing sub-picture will set according to the size of monitoring objective.
3. crop map according to claim 1 is as dense matching method, it is characterized in that, in step S3, described boundary segmentation algorithm is canny boundary segmentation algorithm.
4. crop map is as a dense matching system, it is characterized in that: comprise thin client, the camera be connected with thin client, an AP, also comprises the supply module for thin client, an AP power, and server, the 2nd AP that is connected with server.
5. crop map according to claim 4 is as dense matching system, it is characterized in that: described supply module comprises controller, and the accumulator, solar panels, the inverter that are connected with controller respectively, described inverter is connected with thin client, an AP respectively.
6. crop map according to claim 4 is as dense matching system, it is characterized in that: described camera is two Logitch Pro9000 cameras, and two camera distance 6 ~ 12 centimetres, ground, two camera directions are consistent.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510111324.7A CN104751455A (en) | 2015-03-13 | 2015-03-13 | Crop image dense matching method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510111324.7A CN104751455A (en) | 2015-03-13 | 2015-03-13 | Crop image dense matching method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104751455A true CN104751455A (en) | 2015-07-01 |
Family
ID=53591074
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510111324.7A Pending CN104751455A (en) | 2015-03-13 | 2015-03-13 | Crop image dense matching method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104751455A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106548482A (en) * | 2016-10-19 | 2017-03-29 | 成都西纬科技有限公司 | It is a kind of based on sparse matching and the dense matching method and system of image border |
CN107545567A (en) * | 2017-07-31 | 2018-01-05 | 中国科学院自动化研究所 | The method for registering and device of biological tissue's sequence section micro-image |
CN108986150A (en) * | 2018-07-17 | 2018-12-11 | 南昌航空大学 | A kind of image light stream estimation method and system based on non-rigid dense matching |
CN110620910A (en) * | 2019-09-24 | 2019-12-27 | 中国船舶重工集团公司第七0七研究所 | Image information transmission method of dual-camera network transmission system based on OpenCV |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101958926A (en) * | 2010-09-07 | 2011-01-26 | 上海交通大学 | Field image and environment information remote automatic acquisition and transmission device |
US20140092244A1 (en) * | 2012-09-29 | 2014-04-03 | Nec (China) Co., Ltd. | Object search method, search verification method and apparatuses thereof |
CN104036494A (en) * | 2014-05-21 | 2014-09-10 | 浙江大学 | Fast matching computation method used for fruit picture |
-
2015
- 2015-03-13 CN CN201510111324.7A patent/CN104751455A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101958926A (en) * | 2010-09-07 | 2011-01-26 | 上海交通大学 | Field image and environment information remote automatic acquisition and transmission device |
US20140092244A1 (en) * | 2012-09-29 | 2014-04-03 | Nec (China) Co., Ltd. | Object search method, search verification method and apparatuses thereof |
CN104036494A (en) * | 2014-05-21 | 2014-09-10 | 浙江大学 | Fast matching computation method used for fruit picture |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106548482A (en) * | 2016-10-19 | 2017-03-29 | 成都西纬科技有限公司 | It is a kind of based on sparse matching and the dense matching method and system of image border |
CN107545567A (en) * | 2017-07-31 | 2018-01-05 | 中国科学院自动化研究所 | The method for registering and device of biological tissue's sequence section micro-image |
CN107545567B (en) * | 2017-07-31 | 2020-05-19 | 中国科学院自动化研究所 | Registration method and device for biological tissue sequence section microscopic image |
CN108986150A (en) * | 2018-07-17 | 2018-12-11 | 南昌航空大学 | A kind of image light stream estimation method and system based on non-rigid dense matching |
CN108986150B (en) * | 2018-07-17 | 2020-05-22 | 南昌航空大学 | Image optical flow estimation method and system based on non-rigid dense matching |
CN110620910A (en) * | 2019-09-24 | 2019-12-27 | 中国船舶重工集团公司第七0七研究所 | Image information transmission method of dual-camera network transmission system based on OpenCV |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106251399B (en) | A kind of outdoor scene three-dimensional rebuilding method and implementing device based on lsd-slam | |
CN105245841B (en) | A kind of panoramic video monitoring system based on CUDA | |
CN105809640B (en) | Low illumination level video image enhancement based on Multi-sensor Fusion | |
CN104036488B (en) | Binocular vision-based human body posture and action research method | |
US8005264B2 (en) | Method of automatically detecting and tracking successive frames in a region of interesting by an electronic imaging device | |
CN105716539B (en) | A kind of three-dimentioned shape measurement method of quick high accuracy | |
KR20180054487A (en) | Method and device for processing dvs events | |
CN108805906A (en) | A kind of moving obstacle detection and localization method based on depth map | |
CN104463859B (en) | A kind of real-time video joining method based on tracking specified point | |
CN107316275A (en) | A kind of large scale Microscopic Image Mosaicing algorithm of light stream auxiliary | |
WO2019075948A1 (en) | Pose estimation method for mobile robot | |
CN104751455A (en) | Crop image dense matching method and system | |
CN105631859B (en) | Three-degree-of-freedom bionic stereo visual system | |
CN206611521U (en) | A kind of vehicle environment identifying system and omni-directional visual module based on multisensor | |
CN106599806A (en) | Local curved-surface geometric feature-based human body action recognition method | |
Chen et al. | Bionic mosaic method of panoramic image based on compound eye of fly | |
CN108897342A (en) | For the positioning and tracing method and system of the civilian multi-rotor unmanned aerial vehicle fast moved | |
CN115375842A (en) | Plant three-dimensional reconstruction method, terminal and storage medium | |
CN112801184A (en) | Cloud tracking method, system and device | |
CN108230242A (en) | A kind of conversion method from panorama laser point cloud to video flowing | |
CN105516661B (en) | Principal and subordinate's target monitoring method that fisheye camera is combined with ptz camera | |
CN109883433A (en) | Vehicle positioning method in structured environment based on 360 degree of panoramic views | |
Ju et al. | Design and experiment of an adaptive cruise weeding robot for paddy fields based on improved YOLOv5 | |
CN110111368B (en) | Human body posture recognition-based similar moving target detection and tracking method | |
CN117156298A (en) | Low-speed small target perception method and system based on artificial insect composite bionics |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150701 |
|
WD01 | Invention patent application deemed withdrawn after publication |