CN103116747A - Method and system for automatically recognizing images of stalks and leaves of corns - Google Patents
Method and system for automatically recognizing images of stalks and leaves of corns Download PDFInfo
- Publication number
- CN103116747A CN103116747A CN2013100769813A CN201310076981A CN103116747A CN 103116747 A CN103116747 A CN 103116747A CN 2013100769813 A CN2013100769813 A CN 2013100769813A CN 201310076981 A CN201310076981 A CN 201310076981A CN 103116747 A CN103116747 A CN 103116747A
- Authority
- CN
- China
- Prior art keywords
- chain code
- code string
- image
- point
- straight
- 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.)
- Granted
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a method for recognizing stalks and leaves of corns. The method includes steps of (1), acquiring an RGB (red, green, blue) color image; (2), converting the RGB color image into a grayscale image; (3), preprocessing the grayscale image; (4), converting the preprocessed grayscale image into a binary image by means of binarization processing; (5), performing straight line detection for the binary image; and (6), computing shape features of different zones, and recognizing the acquired RGB color image according to the shape features of the different zones to finally acquire a stalk target. The different zones are acquired by means of detecting the binary image and are separated from one another by straight lines. The method has the advantages that stalks of the corns can be accurately recognized in real time, and problems of damaging equipment or hurting corn plants and the like due to blockage of stalks when an automatic fertilizer applicator is used for fertilization operation can be prevented.
Description
Technical field
The present invention relates to technical field of image processing, relate in particular to a kind of corn stem recognition methods.
Background technology
In the corn fertilizing operation process, fertilizer apparatus on automatic fertilizer spreaders can bump with " barriers " such as stem stalks of corn, cause device damage or injury crop, therefore need to carry out in real time, identify accurately and locate these " barrier ", make automatic fertilizer spreaders change mobile alignment or stop mobile before may bumping with " barrier ".
This identification and location technology can realize by machine vision, adopt the camera acquisition image, utilize the equipment such as computing machine, dsp chip that image is processed.Available technology adopting is carried out hough conversion straight-line detection to the aberration R-B image of corn stem and then is realized the identification of stem stalk, although the line detection method based on the Hough conversion is a kind of algorithm that reaches its maturity, but calculated amount is larger, is unfavorable for having lost in application in real time and conversion process end points and the length information of line segment.
Summary of the invention
The technical problem to be solved in the present invention is: a kind of corn stem recognition methods is provided, it can identify the stem stalk of corn in real time, exactly, and can make automatic fertilizer spreaders avoid generation because of problems such as the obstruction cause damage of equipment of stem stalk or injury milpas during operation in fertilising.
For addressing the above problem, the invention provides a kind of corn recognition methods, comprise the following steps:
(1) gather the RGB coloured image;
(2) described RGB coloured image is converted into gray level image;
(3) described gray level image is carried out pre-service;
(4) by binary conversion treatment, pretreated described gray level image is converted into binary image;
(5) described binary image is carried out straight-line detection;
(6) calculate the shape facility of the zones of different of the detected line segmentation that obtains of described binary image, and according to the shape facility of described zones of different, the RGB coloured image that collects is identified, obtain final stem stalk target.
Described pre-service is that Lifting Wavelet is processed, and it can extract gray level image low frequency profile information, suppress high frequency noise.
When the pre-service in step (3) was the Lifting Wavelet processing, step (4) further comprised:
Gray level image after pixel after processing through Lifting Wavelet is reduced adopts differentiating operator to carry out binary conversion treatment, and gray level image is converted into binary image.
Step (5) further comprises:
Utilize zero crossing to detect the frontier point that obtains binary image, and then by edge following algorithm, frontier point is converted into the symbolic information of chain representation, follow the tracks of based on the border chain code string that obtains and carry out straight-line detection.
After the present invention carries out gray processing, Lifting Wavelet, binary conversion treatment by the RGB coloured image to corn stem, pass through boundary points detection, utilization is based on the straight-line detection of chain code and then can identify in real time, exactly the stem stalk of corn, makes automatic fertilizer spreaders avoid generation because of problems such as the obstruction cause damage of equipment of stem stalk or injury milpas when the fertilising operation.
Description of drawings
Fig. 1 is the process flow diagram of corn stem recognition methods described in embodiment of the present invention;
Fig. 2 is the schematic diagram of 8 direction chain codes.
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 explanation the present invention, but are not used for limiting the scope of the invention.
As shown in Figure 1, a kind of corn stem recognition methods of the present invention comprises the following steps:
(1) gather the RGB coloured image;
(2) described RGB coloured image is converted into gray level image;
In this step, the RGB coloured image that collects is carried out gray processing process, coloured image is converted into gray level image.If in coloured image, the color of original certain point is RGB (R, G, B), can pass through following several method, be converted into gray scale:
1) floating-point arithmetic: Gray=R*0.3+G*0.59+B*0.11;
2) integer method: Gray=(R*30+G*59+B*11)/100;
3) displacement method: Gray=(R*76+G*151+B*28)〉〉 8;
4) mean value method: Gray=(R+G+B)/3;
5) only get green: Gray=G;
After trying to achieve Gray by above-mentioned any method, R, G in original RGB (R, G, B), B unification are replaced with Gray, form new color RGB (Gray, Gray, Gray), replace original RGB (R, G, B) with it and just obtained gray level image.
In image processing process, for the color characteristic that utilizes coloured image splits destination object from image background, often coloured image is carried out dimension-reduction treatment, only utilize a certain characteristic component to come the color of Description Image, coloured image is converted into gray level image.Meanwhile, by the gray processing of coloured image, image data amount is obviously reduced, greatly improved image processing speed.In the present invention, because the color information of image is unimportant to the culm morphology information that will extract, gray level image has kept the shape information of stem stalk in the original color image preferably.
(3) described gray level image is carried out pre-service;
Above-mentioned pre-service is that Lifting Wavelet is processed, and the gray level image after transforming is carried out Lifting Wavelet process, and described Lifting Wavelet is processed can extract gray level image low frequency profile information, suppress high frequency noise;
Lifting Wavelet is a kind of implementation method of wavelet transformation more fast and effectively, is called as Second Generation Wavelet Transformation.Lifting Wavelet does not rely on Fourier transform, has inherited the feature of the multiresolution of first generation small echo, and the coefficient after wavelet transformation is integer, need not extra internal memory, and computing is carried out in available bit manipulation, can realize the wavelet transformation of arbitrary image size.By means of the factorization wavelet transformation, all wavelet transformations can both be realized with the Lifting Wavelet pattern.
In this step, Lifting Wavelet is processed and is comprised division, prediction and upgrade, and considers signal s
j={ s
j,l| 0≤l≤2
j, the low frequency signal that obtains after its process one-level wavelet transformation is s
j-1With high-frequency signal d
j-1, the specific implementation process that builds low-resolution image is:
I. division: with input signal s
jBe divided into strange, even two subsets, the set that is positioned at the element formation of even subscript position is designated as even
j-1={ s
J, 2l| 0≤l≤2
j-1-1}, the set that is positioned at the element formation of strange subscript position is designated as odd
j-1={ s
J, 2l+1| 0≤l≤2
j-1-1};
II. prediction: on the basis based on the raw data correlativity, go prediction or interpolation odd number sequence with the predicted value of even number sequence;
III. upgrade: the purpose of renewal is to look for a better subset s
j-1, make its a certain scalar characteristic Q (x) (as constant in root-mean-square value) that keeps former figure, i.e. Q (s
j-1)=Q (s
j);
Wherein, for Lifting Wavelet, there are minute split operator Split, predictive operator P and upgrade operator U, making
(even
j-1，odd
j-1)=Split(s
j),
d
j-1=odd
j-1-P(even
j-1),
s
j-1=even
j-1+U(d
j-1)。
One width pixel is the image of M * N, through after Lifting Wavelet, gets its low frequency component and builds low-resolution image, and its pixel is
In addition, mostly be distributed in the HFS of image due to noise, therefore effectively suppressed high frequency noise based on the constructed low-resolution image of Lifting Wavelet.
(4) by binary conversion treatment, pretreated described gray level image is converted into binary image;
When the pre-service in step (3) was the Lifting Wavelet processing, this step further comprised: the gray level image after the pixel after processing through Lifting Wavelet is reduced adopts differentiating operator to carry out binary conversion treatment, and gray level image is converted into binary image;
(5) described binary image is carried out straight-line detection;
This step further comprises:
Utilize zero crossing to detect the frontier point that obtains binary image, and then by edge following algorithm, frontier point is converted into the symbolic information of chain representation, follow the tracks of based on the border chain code string that obtains and carry out straight-line detection;
in this step, edge following algorithm mainly comprises scanning process and tracing process, hypothetical boundary is present between pixel, unit consists of by the limit, be divided into two kinds of horizontal and verticals, if current investigation point coordinate is (x, y), judge whether to exist horizontal sides unit with (x+1, y) be reference point, judge whether to exist perpendicular edge unit with (x, y+1) be reference point, when the value of investigating point and reference point is identical, do not have the border to pass through between these two points, the value of limit unit is zero, all that impact point or background dot are divided into two kinds of positive zero-sum negative zeros according to them, there is the border to pass through between 2 when the value of investigating point and reference point is different, the value non-zero of limit unit, that impact point or background dot are got respectively plus or minus according to investigating point,
Tracking direction provides as follows: impact point is all the time on the left side of working direction, outer boundary for a connected region, tracking direction is counterclockwise, inner boundary (the formed border of inner void) is clockwise direction, when quadraturing on sealing chain code basis, the region area that the outer boundary chain code surrounds is being for just, and zone that the inner boundary chain code surrounds and long-pending for negative;
The value of limit unit has also shown corresponding tracking direction simultaneously, and the first direction in certain limit is for just, and for horizontal sides unit, tracking direction to the right, and is first for perpendicular edge, and tracking direction upwards;
Whether mark in tracing process is beaten on the impact point of front unit, to distinguish this frontier point tracked mistake;
When having non-zero limit unit between current investigation point and reference point, according to all the other situations of 2 in these 22 * 2 windows that consist of, decide next step tracking direction, i.e. follow-up limit unit, when return to origin, tracing process finishes.
Wherein, scanning process adopts the mode of network scanning to seek the frontier point of not yet following the tracks of, and chain code adopts 8 direction chain codes.Above-mentioned border is followed the tracks of core algorithm and has greatly been improved efficiency of algorithm.
Image is done network scanning, only have the zone of intersecting with mesh lines just possible tracked, those zones of dropping on fully in grid are left in the basket.Therefore, adopt network scanning also can greatly improve processing speed.
Follow the tracks of based on the border chain code string that obtains and carry out in the straight-line detection process, adopt following form to represent the chain code string of object boundary:
I：Len,X,Y,d
1,d
2,…,d
n
Wherein I is the numbering of current chain code string, and Len is the total length of chain code in current chain code string, and X, Y are the image coordinate that current chain code strings initial point, d
1, d
2..., d
nBe the direction code of each pixel on current chain code string, the step of straight-line detection is:
IV. minimum length of straigh line checking: order judges each chain code string, if the length of chain code string is given up this chain code string less than default minimum length of straigh line threshold value LT;
V. extract straight-line segment according to straight-line segment degree of approximation criterion from the chain code string:
Order judges each chain code string;
1. from the off, select successively to satisfy the subchain code string of minimum length of straigh line constraint, calculate the straight-line segment degree of approximation S of this section according to formula (1), if S 〉=ST, ST is default minimum straight-line segment similarity threshold, and this section satisfies line constraint, turns 2., otherwise give up this section, continue next cross-talk chain code string of judgement;
Wherein, || p
s-p
e|| expression end points p
sWith p
eBetween the ideal line distance, Len (p
s, p
e) represent that subchain code string is from end points p
sTo end points p
ePhysical length through pixel, S represents the straight-line segment degree of approximation, in addition, when calculating the physical length of chain code string, if direction code is 0,2,4,6, the physical length between two pixels of chain code connection is 1, if direction code is 1,3,5,7, the physical length between two pixels is that 2(is referring to Fig. 2);
If 2. the last period, subchain code string also satisfied line constraint, consider that can two adjacent strip chain code strings be merged into a longer straight-line segment, the starting point of chain code string is regarded a new chain code string as to this section chain code between the terminal point of this section chain code string with the last period, calculates its straight-line segment degree of approximation S '.If S ' 〉=ST merges, otherwise this cross-talk chain code string is marked as a new straight-line segment;
If 3. oneself arrives chain code string terminal point, finish, 1. continue next cross-talk chain code string of judgement otherwise turn.
Wherein, described default minimum length of straigh line threshold value LT is 23 pixels, and default minimum straight-line segment similarity threshold ST is 0.91.
The present invention first carries out chain code following to object boundary, then carries out line segment extraction in the chain code set of strings that obtains, and its advantage is that calculated amount is little, and can obtain simultaneously the information such as position, length, direction of straight-line segment.The line detection algorithm that the present invention adopts has been broken away from the constraint of Freeman ideal line criterion from practical application, only adopts minimum length of straigh line and two constrained parameters of minimum line similar degree, can realize rapidly and accurately straight-line detection.This algorithm detection speed is fast, practical, has strengthened the real-time of corn stem identifying.
(6) calculate the shape facility of the zones of different of the detected line segmentation that obtains of described binary image, and according to the shape facility of described zones of different, the RGB coloured image that collects is identified, obtain final stem stalk target.
In this step, described shape facility has a plurality of parameters, and the present embodiment only calculates form parameter and excentricity.The straight line region is identified as the stem stalk, to being utilized form parameter by other zone of line segmentation
And excentricity
Identify, satisfy F greater than 2.3 and e be identified as the step of stem stalk greater than 5 zone, wherein, Area and Per are respectively the area and perimeter in zone, c and a are respectively border long axis length and the minor axis length in zone, major axis refers on the zone boundary distance 2 lines farthest, on the zone boundary in the line vertical with major axis the longest line segment be called minor axis.
Above embodiment only is used for explanation the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; without departing from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (10)
1. a method of automatically identifying the corn stem leaf image, is characterized in that, comprises the following steps:
(1) gather the RGB coloured image;
(2) described RGB coloured image is converted into gray level image;
(3) described gray level image is carried out pre-service;
(4) by binary conversion treatment, pretreated described gray level image is converted into binary image;
(5) described binary image is carried out straight-line detection;
(6) calculate the shape facility of the zones of different of the detected line segmentation that obtains of described binary image, and according to the shape facility of described zones of different, the RGB coloured image that collects is identified, obtain final stem stalk target.
2. the method for claim 1, described pre-service are that Lifting Wavelet is processed, and it can extract gray level image low frequency profile information, suppress high frequency noise.
3. method as claimed in claim 2, described Lifting Wavelet are processed and are comprised division, prediction and upgrade, consideration signal s
j={ s
j,l| 0≤l≤2
j, the low frequency signal that obtains after its process one-level wavelet transformation is s
j-1With high-frequency signal d
j-1, the specific implementation process that builds low-resolution image is:
I. division: with input signal s
jBe divided into strange, even two subsets, the set that is positioned at the element formation of even subscript position is designated as even
j-1={ s
J, 2l| 0≤l≤2
j-1-1}, the set that is positioned at the element formation of strange subscript position is designated as odd
j-1={ s
J, 2l+1| 0≤l≤2
j-1-1};
II. prediction: on the basis based on the raw data correlativity, go prediction or interpolation odd number sequence with the predicted value of even number sequence;
III. upgrade: the purpose of renewal is to look for a better subset s
j-1, make its a certain scalar characteristic Q (x) (as constant in root-mean-square value) that keeps former figure, i.e. Q (s
j-1)=Q (s
j);
Wherein, for Lifting Wavelet, there are minute split operator Split, predictive operator P and upgrade operator U, making
(even
j-1，odd
j-1)=Split(s
j),
d
j-1=odd
j-1-P(even
j-1),
s
j-1=even
j-1+U(d
j-1)。
4. method as claimed in claim 3, a certain scalar characteristic Q (x) of the former figure of wherein said maintenance is that root-mean-square value is constant.
5. as claim 2,3 or 4 described methods, step (4) further comprises:
Gray level image after pixel after processing through Lifting Wavelet is reduced adopts differentiating operator to carry out binary conversion treatment, and gray level image is converted into binary image.
6. the method for claim 1, step (5) further comprises:
Utilize zero crossing to detect the frontier point that obtains binary image, and then by edge following algorithm, frontier point is converted into the symbolic information of chain representation, follow the tracks of based on the border chain code string that obtains and carry out straight-line detection.
7. method as claimed in claim 6, described edge following algorithm mainly comprises scanning process and tracing process, hypothetical boundary is present between pixel, unit consists of by the limit, be divided into two kinds of horizontal and verticals, if current investigation point coordinate is (x, y), judge whether to exist horizontal sides unit with (x+1, y) be reference point, judge whether to exist perpendicular edge unit with (x, y+1) be reference point, when the value of investigating point and reference point is identical, do not have the border to pass through between these two points, the value of limit unit is zero, all that impact point or background dot are divided into two kinds of positive zero-sum negative zeros according to them, there is the border to pass through between 2 when the value of investigating point and reference point is different, the value non-zero of limit unit, that impact point or background dot are got respectively plus or minus according to investigating point,
Tracking direction provides as follows: impact point is all the time on the left side of working direction, outer boundary for a connected region, tracking direction is counterclockwise, inner boundary is clockwise direction, when quadraturing on sealing chain code basis, the region area that the outer boundary chain code surrounds is being for just, and zone that the inner boundary chain code surrounds and long-pending for negative;
The value of limit unit has also shown corresponding tracking direction simultaneously, and the first direction in certain limit is for just, and for horizontal sides unit, tracking direction to the right, and is first for perpendicular edge, and tracking direction upwards;
Whether mark in tracing process is beaten on the impact point of front unit, to distinguish this frontier point tracked mistake;
When having non-zero limit unit between current investigation point and reference point, according to all the other situations of 2 in these 22 * 2 windows that consist of, decide next step tracking direction, i.e. follow-up limit unit, when return to origin, tracing process finishes.
8. method as claimed in claim 7, described scanning process adopt the mode of network scanning to seek the frontier point of not yet following the tracks of, and described chain code is 8 direction chain codes, and described inner boundary is the formed border of inner void.
9. method as claimed in claim 6 is describedly followed the tracks of based on the border chain code string that obtains and is carried out in the straight-line detection process, adopts following form to represent the chain code string of object boundary:
I：Len,X,Y,d
1,d
2,…,d
n
Wherein I is the numbering of current chain code string, and Len is the total length of chain code in current chain code string, and X, Y are the image coordinate that current chain code strings initial point, d
1, d
2..., d
nBe the direction code of each pixel on current chain code string, the step of straight-line detection is:
IV. minimum length of straigh line checking: order judges each chain code string, if the length of chain code string is given up this chain code string less than default minimum length of straigh line threshold value LT;
V. extract straight-line segment according to straight-line segment degree of approximation criterion from the chain code string:
Order judges each chain code string;
1. from the off, select successively to satisfy the subchain code string of minimum length of straigh line constraint, calculate the straight-line segment degree of approximation S of this section according to formula (1), if S 〉=ST, ST is default minimum straight-line segment similarity threshold, and this section satisfies line constraint, turns 2., otherwise give up this section, continue next cross-talk chain code string of judgement;
Wherein, || p
s-p
e|| expression end points p
sWith p
eBetween the ideal line distance, Len (p
s, p
e) represent that subchain code string is from end points p
sTo end points p
eThrough the physical length of pixel, S represents the straight-line segment degree of approximation, in addition, when calculating the physical length of chain code string, if direction code is 0,2,4,6, the physical length between two pixels of chain code connection is 1, if direction code is 1,3,5,7, the physical length between two pixels is 2;
If 2. the last period, subchain code string also satisfied line constraint, consider that can two adjacent strip chain code strings be merged into a longer straight-line segment, the starting point of chain code string is regarded a new chain code string as to this section chain code between the terminal point of this section chain code string with the last period, calculates its straight-line segment degree of approximation S '.If S ' 〉=ST merges, otherwise this cross-talk chain code string is marked as a new straight-line segment;
If 3. oneself arrives chain code string terminal point, finish, 1. continue next cross-talk chain code string of judgement otherwise turn;
Wherein said default minimum length of straigh line threshold value LT is 23 pixels, and default minimum straight-line segment similarity threshold ST is 0.91.
10. the method for claim 1, described step (6) further comprises: the straight line region is identified as the stem stalk, to being utilized form parameter by other zone of line segmentation
And excentricity
Identify, satisfy F greater than 2.3 and e be identified as the step of stem stalk greater than 5 zone, wherein, Area and Per be respectively the zone area and perimeter, c and a be respectively the zone border long axis length and minor axis length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310076981.3A CN103116747B (en) | 2013-03-11 | 2013-03-11 | Automatically the method and system of corn stem leaf image is identified |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310076981.3A CN103116747B (en) | 2013-03-11 | 2013-03-11 | Automatically the method and system of corn stem leaf image is identified |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103116747A true CN103116747A (en) | 2013-05-22 |
| CN103116747B CN103116747B (en) | 2016-06-08 |
Family
ID=48415119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310076981.3A Expired - Fee Related CN103116747B (en) | 2013-03-11 | 2013-03-11 | Automatically the method and system of corn stem leaf image is identified |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103116747B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103714342B (en) * | 2013-12-20 | 2016-10-05 | 华北电力大学（保定） | Insulator chain automatic positioning method of taking photo by plane based on bianry image shape facility |
| CN107734120A (en) * | 2016-08-10 | 2018-02-23 | 北京东舟技术股份有限公司 | The fluency method for testing performance of terminal device |
| CN107941802A (en) * | 2017-10-31 | 2018-04-20 | 华中农业大学 | Pot rice leaf roll degree measurement method based on machine vision |
| CN109886094A (en) * | 2019-01-08 | 2019-06-14 | 中国农业大学 | A kind of crop growth of cereal crop seedlings seedling gesture capturing analysis method and device |
| CN112286230A (en) * | 2020-11-13 | 2021-01-29 | 国家电网有限公司 | Unmanned aerial vehicle visual image algorithm, obstacle avoidance step and information fusion processing system thereof |
| CN113807129A (en) * | 2020-06-12 | 2021-12-17 | 广州极飞科技股份有限公司 | Crop area identification method and device, computer equipment and storage medium |
| CN114972979A (en) * | 2022-01-13 | 2022-08-30 | 江苏省农业科学院 | Plant main stem detection method used in greenhouse environment snapshot picture |
| CN114972978A (en) * | 2022-01-13 | 2022-08-30 | 江苏省农业科学院 | Method for detecting main stem of plant in seedling stage in phenotypic imaging box environment |
| CN115147423A (en) * | 2022-09-06 | 2022-10-04 | 江苏欧罗曼家纺有限公司 | Cotton top detection method based on improved chain code method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101645172A (en) * | 2009-09-09 | 2010-02-10 | 北京理工大学 | Rapid detection method for straight line in digital image |
| US20100061629A1 (en) * | 2008-09-05 | 2010-03-11 | Digital Business Processes, Inc. | Method and Apparatus for Binarization Threshold Calculation |
-
2013
- 2013-03-11 CN CN201310076981.3A patent/CN103116747B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100061629A1 (en) * | 2008-09-05 | 2010-03-11 | Digital Business Processes, Inc. | Method and Apparatus for Binarization Threshold Calculation |
| CN101645172A (en) * | 2009-09-09 | 2010-02-10 | 北京理工大学 | Rapid detection method for straight line in digital image |
Non-Patent Citations (5)
| Title |
|---|
| 史册: "对一种快速边缘跟踪算法的讨论", 《小型微型计算机系统》 * |
| 孙涵等: "一种快速实用的直线检测算法", 《计算机应用研究》 * |
| 牛珂等: "基于图像处理的植物叶片参数测量系统", 《农村经济与科技》 * |
| 葛雯等: "一种基于小波提升变换的多尺度边缘提取算法", 《东北大学学报（自然科学版）》 * |
| 赖军臣: "基于病症图像的玉米病害智能诊断研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103714342B (en) * | 2013-12-20 | 2016-10-05 | 华北电力大学（保定） | Insulator chain automatic positioning method of taking photo by plane based on bianry image shape facility |
| CN107734120A (en) * | 2016-08-10 | 2018-02-23 | 北京东舟技术股份有限公司 | The fluency method for testing performance of terminal device |
| CN107941802A (en) * | 2017-10-31 | 2018-04-20 | 华中农业大学 | Pot rice leaf roll degree measurement method based on machine vision |
| CN107941802B (en) * | 2017-10-31 | 2020-05-08 | 华中农业大学 | Potted rice leaf rolling degree measuring method based on machine vision |
| CN109886094A (en) * | 2019-01-08 | 2019-06-14 | 中国农业大学 | A kind of crop growth of cereal crop seedlings seedling gesture capturing analysis method and device |
| CN113807129A (en) * | 2020-06-12 | 2021-12-17 | 广州极飞科技股份有限公司 | Crop area identification method and device, computer equipment and storage medium |
| CN112286230A (en) * | 2020-11-13 | 2021-01-29 | 国家电网有限公司 | Unmanned aerial vehicle visual image algorithm, obstacle avoidance step and information fusion processing system thereof |
| CN114972979A (en) * | 2022-01-13 | 2022-08-30 | 江苏省农业科学院 | Plant main stem detection method used in greenhouse environment snapshot picture |
| CN114972978A (en) * | 2022-01-13 | 2022-08-30 | 江苏省农业科学院 | Method for detecting main stem of plant in seedling stage in phenotypic imaging box environment |
| CN115147423A (en) * | 2022-09-06 | 2022-10-04 | 江苏欧罗曼家纺有限公司 | Cotton top detection method based on improved chain code method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103116747B (en) | 2016-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103116747A (en) | Method and system for automatically recognizing images of stalks and leaves of corns | |
| CN103400151B (en) | The optical remote sensing image of integration and GIS autoregistration and Clean water withdraw method | |
| CN107895367B (en) | Bone age identification method and system and electronic equipment | |
| CN102194114B (en) | Method for recognizing iris based on edge gradient direction pyramid histogram | |
| CN104008551B (en) | A kind of Citrus Huanglongbing pathogen detection method based on visible images | |
| CN112200884B (en) | Lane line generation method and device | |
| CN104268515A (en) | Sperm morphology anomaly detection method | |
| CN103177261A (en) | Image-recognition-technology-based cow milk yield auto-metering system and image recognition method therefor | |
| Liu et al. | A lightweight method for grape berry counting based on automated 3D bunch reconstruction from a single image | |
| CN112541383B (en) | Method and device for identifying weed area | |
| Liang et al. | Segmentation and weight prediction of grape ear based on SFNet-ResNet18 | |
| CN103530606A (en) | Agricultural machine navigation path extraction method under weed environment | |
| CN107784469A (en) | A kind of articles from the storeroom checking method based on unmanned plane image technique | |
| CN104952070A (en) | Near-rectangle guide based remote-sensing cornfield image segmentation method | |
| CN113920148B (en) | Building boundary extraction method and equipment based on polygon and storage medium | |
| CN114092378A (en) | Animal health detection method, device, equipment and storage medium | |
| CN116433668A (en) | Intelligent hydraulic oil pipe oil leakage detection method | |
| CN116452667A (en) | Target identification and positioning method based on image processing | |
| CN114758132B (en) | Fruit tree disease and pest identification method and system based on convolutional neural network | |
| CN106909881A (en) | The method and system of corn breeding base ridge number are extracted based on unmanned aerial vehicle remote sensing images | |
| CN101853503A (en) | Spectral line inflexion multi-scale optimizing segmentation method and application thereof | |
| CN104298993A (en) | Pole number positioning and identification method suitable for railway line in complex scene | |
| CN117349714B (en) | Classification method, system, equipment and medium for medical image of Alzheimer disease | |
| CN112016391B (en) | Fishpond identification method, system and medium based on high-resolution satellite remote sensing image | |
| CN110060294A (en) | A kind of yield assessment method of fruit tree crop |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160608 Termination date: 20170311 |