CN106447685B - A kind of infrared track method - Google Patents
A kind of infrared track method Download PDFInfo
- Publication number
- CN106447685B CN106447685B CN201610803370.8A CN201610803370A CN106447685B CN 106447685 B CN106447685 B CN 106447685B CN 201610803370 A CN201610803370 A CN 201610803370A CN 106447685 B CN106447685 B CN 106447685B
- Authority
- CN
- China
- Prior art keywords
- point
- template
- target
- edge
- frame image
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000003044 adaptive effect Effects 0.000 claims abstract description 25
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 abstract description 12
- 238000011160 research Methods 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
Classifications
-
- 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
- G06T2207/10048—Infrared image
Landscapes
- Image Analysis (AREA)
Abstract
The infrared track method that the embodiment of the invention discloses a kind of based on Hausdorff apart from matched adaptive template, steps are as follows: firstly, carrying out edge detection using infrared image of the improved sobel edge detection operator to input, extracting the contour feature of target;Secondly, carrying out template matching to target template and region to be searched using quick two-way Hausdorff distance method in tracking;Finally, carrying out adaptive template renewal to optimal matching result.The present invention can for a long time stablize infrared target, precisely track, and arithmetic speed is fast, have compared with hard real-time and robustness.
Description
Technical field
The present invention relates to digital image processing fields, are based on Hao Siduofu (Hausdorff) distance more particularly, to one kind
The infrared track method of matched adaptive template.
Background technique
Nowadays, with the continuous promotion of information technology and computer performance, more and more people have started to computer
Target motion problems in vision are studied.In infrared image processing field, infrared object tracking intelligent security-protecting and monitoring with
And the rapid development of infrared guidance technology research field, the exigent prison especially in the real-time of some tracking and precision
In control and operational environment, there is great meaning to the research of new algorithm.
Infrared object tracking is in Engineering Control, traffic monitoring, medical image research, automated navigation system, astronomical monitoring etc.
There is critically important practical value in field.Especially in military aspect, infrared guidance become more and more important main battle weaponry it
One.There is also many shortcomings for the algorithm of current most of infrared object trackings, and more complicated algorithm is in real-time target
Tracking aspect does not reach requirement, single algorithm not can be carried out again it is steady in a long-term track, this is for modernizing information war
Striving is totally unfavorable influence, so the current people research that the infrared object tracking algorithm of research real-time high-efficiency becomes is important
Project.
It apart from matched infrared object tracking algorithm is that contour feature based on infrared target carries out mesh based on Hausdorff
Mark modeling can be very good specific target appearance feature to carry out robust tracking, add adaptive target template more
Newly, influence of the interference to tracking of noise can effectively be reduced.But when current infrared target be blocked, dimensional variation the problems such as
Also need deeper into research.
Summary of the invention
The object of the present invention is to provide a kind of infrared track sides based on Hausdorff apart from matched adaptive template
Method, tracking stabilization good for infrared object tracking real-time and strong robustness.
Technical solution disclosed by the invention are as follows: using improved Sobel edge detection operator to the infrared image of input into
Row edge detection extracts the contour feature of target, in tracking using quick two-way Hausdorff distance method to target mould
Plate and region to be searched carry out template matching, adaptive template renewal is carried out to optimal matching result to carry out target with
Track, the specific steps are as follows:
Step 1: input infrared video carries out artificial selected target region to initial frame, using improved Sobel operator into
Row binary conversion treatment obtains the edge contour information of target and establishes target template;
Step 2: next frame image is read, the Sobel operator that region to be detected improves is handled, after binaryzation,
Obtain edge contour information;
Step 3: to the target template and the two-way Hausdorff distance that improves of region to be detected after Sobel
Matching, finds out the position of best match;
Step 4: judging the update of target template using adaptive mode slat element, target template is updated if meeting condition
And enter next frame, target template is not otherwise updated and enters next frame until Infrared video image terminates.
Further, binary conversion treatment is carried out using improved Sobel operator in step 1, i.e., by edge pixel gray value
It is set as 1, rest of pixels value is set as 0, obtains the edge contour information of target and establishes target template, specific as follows:
To the Sobel edge detection process that the infrared image gray value f (i, j) of input is improved, wherein i, j points become
Image pixel cross, ordinate value, Sobel operator is the edge detection operator of first derivative, and formula is as follows:
Sx=(Z1+2Z2+Z3)-(Z7+2Z8+Z9)
Sy=(Z1+2Z4+Z7)-(Z3+2Z6+Z9)
Wherein, Z1,Z2,Z3,Z4,Z5,Z6,Z7,Z8,Z9The 8 neighborhood territory pixel gray values of respectively pixel f (i, j), Sx,SyPoint
It Wei not horizontal, vertical direction gradient magnitude.
The gradient that improved Sobel operator increases positive and negative 45 ° of directions calculates, and formula is as follows:
S45°=(Z2+2Z3+Z6)-(Z4+2Z7+Z8)
S-45°=(Z2+2Z1+Z4)-(Z6+2Z8+Z9)
Work as Sx,Sy,S45°,S-45°In any one be greater than preset threshold value T when for be edge, record the side of target area
The coordinate value of edge pixel establishes target template.
Further, in step 3 to after Sobel target template and region to be detected improve it is two-way
Hausdorff distance matching, specific as follows:
To the marginal information set A={ a in target template1,a2,...,amAnd region candidate template to be measured edge letter
Cease set B={ b1,b2,...,bnTwo-way Hausdorff distance matching is carried out, formula is as follows:
H (A, B)=max (h (A, B), h (B, A))
Wherein, a1,a2,...,amFor the coordinate information of target template edge aggregation, b1,b2,...,bmFor candidate mould to be measured
The edge coordinate of plate,
H (A, B)=max (a ∈ A) min (b ∈ B) | | a-b | |
H (B, A)=max (b ∈ B) min (a ∈ A) | | b-a | |
H (A, B) indicates the point a from set A in above formula (6)iTo the point that nearest set B is put apart from this distance set into
Row sequence, takes maximum value therein, and h (B, A) indicates the point b from set B in above formulaiTo the point for putting nearest set A apart from this
Distance set be ranked up, take maximum value therein, | | | | be apart from normal form.
When the Hausdorff improved is apart from matching primitives, to current point ai8 contiguous ranges in look for another set
Whether the point of this range is had in coordinate, if aiFour neighborhood positions up and down correspond in set B and have at least one corresponding point,
Then calculate thisiPoint is 1 to set B point Hausdorff distance, otherwise continually looks for aiEight neighborhood point, which corresponds in set B, to be had accordingly
At least one point, then calculate thisiPoint is 2 to set B point Hausdorff distance, and otherwise distance is just denoted as 10.
Further, the update that target template is judged using adaptive mode slat element in step 4, detailed process is as follows:
By calculating the edge point set sum S in target templateAAnd the marginal point of best match position region template
Gather sum SB, calculate Adaptive template-updating conditional parameter P1=SA/SBAnd the gray scale of target template region original image
Mean value hAWith the original image gray average h of best match position region templateB, calculate Adaptive template-updating conditional parameter P2
=hA/hB, and if only if meeting condition α < P1< β and condition γ < P2When < λ, update current region template is target template,
Otherwise it does not update, wherein α, beta, gamma, λ is constant.To reduce the interference of noise, real-time, stable infrared object tracking is carried out.
Detailed description of the invention
Fig. 1 is the stream of the infrared track method based on Hausdorff apart from matched adaptive template of the method for the present invention
Cheng Tu.
Fig. 2 is the Sobel filter template schematic diagram both horizontally and vertically of the method for the present invention.
Fig. 3 is the Sobel filter template schematic diagram in positive and negative 45 degree of directions of the method for the present invention.
Specific embodiment
Below in conjunction with attached drawing the present invention will be described in detail method based on Hausdorff apart from matched adaptive template
Infrared track method specific implementation process.
As shown in Figure 1, in the method for the present invention, it is a kind of based on Hausdorff apart from matched adaptive template it is infrared with
Track method carries out edge detection using infrared image of the improved Sobel edge detection operator to input, extracts the profile of target
Feature carries out template to target template and region to be searched using quick two-way Hausdorff distance method in tracking
Match, adaptive template renewal carried out to carry out target following to optimal matching result, the specific steps are as follows:
Step 1: input infrared video carries out artificial selected target region to initial frame, using improved Sobel operator into
Row binary conversion treatment obtains the edge contour information of target and establishes target template;
Step 2: next frame image is read, the Sobel operator that region to be detected improves is handled, after binaryzation,
Obtain edge contour information;
Step 3: to the target template and the two-way Hausdorff distance that improves of region to be detected after Sobel
Matching, finds out the position of best match;
Step 4: judging the update of target template using adaptive mode slat element, target template is updated if meeting condition
And enter next frame, target template is not otherwise updated and enters next frame until Infrared video image terminates.
Further, binary conversion treatment is carried out using improved Sobel operator in step 1, i.e., by edge pixel gray value
It is set as 1, rest of pixels value is set as 0, obtains the edge contour information of target and establishes target template, specific as follows:
To the Sobel edge detection process that the infrared image gray value f (i, j) of input is improved, wherein i, j points become
Image pixel cross, ordinate value, Sobel operator is the edge detection operator of first derivative, and formula is as follows:
Sx=(Z1+2Z2+Z3)-(Z7+2Z8+Z9)
Sy=(Z1+2Z4+Z7)-(Z3+2Z6+Z9)
Wherein, Z1,Z2,Z3,Z4,Z5,Z6,Z7,Z8,Z9The 8 neighborhood territory pixel gray values of respectively pixel f (i, j), Sx,SyPoint
It Wei not horizontal, vertical direction gradient magnitude.
The gradient that improved Sobel operator increases positive and negative 45 ° of directions calculates, and formula is as follows:
S45°=(Z2+2Z3+Z6)-(Z4+2Z7+Z8)
S-45°=(Z2+2Z1+Z4)-(Z6+2Z8+Z9)
Work as Sx,Sy,S45°,S-45°In any one be greater than preset threshold value T when for be edge, record the side of target area
The coordinate value of edge pixel establishes target template.
Further, in step 3 to after Sobel target template and region to be detected improve it is two-way
Hausdorff distance matching, specific as follows:
To the marginal information set A={ a in target template1,a2,...,amAnd region candidate template to be measured edge letter
Cease set B={ b1,b2,...,bnTwo-way Hausdorff distance matching is carried out, formula is as follows:
H (A, B)=max (h (A, B), h (B, A))
Wherein, a1,a2,...,amFor the coordinate information of target template edge aggregation, b1,b2,...,bmFor candidate mould to be measured
The edge coordinate of plate,
H (A, B)=max (a ∈ A) min (b ∈ B) | | a-b | |
H (B, A)=max (b ∈ B) min (a ∈ A) | | b-a | |
H (A, B) indicates the point a from set A in above formulaiDistance set to the point for putting nearest set B apart from this carries out
Sequence, takes maximum value therein, and h (B, A) indicates the point b from set B in above formulaiTo the point for putting nearest set A apart from this
Distance set is ranked up, and takes maximum value therein, | | | | it is apart from normal form.
When the Hausdorff improved is apart from matching primitives, to current point ai8 contiguous ranges in look for another set
Whether the point of this range is had in coordinate, if aiFour neighborhood positions up and down correspond in set B and have at least one corresponding point,
Then calculate thisiPoint is 1 to set B point Hausdorff distance, otherwise continually looks for aiEight neighborhood point, which corresponds in set B, to be had accordingly
At least one point, then calculate thisiPoint is 2 to set B point Hausdorff distance, and otherwise distance is just denoted as 10.
Further, the update that target template is judged using adaptive mode slat element in step 4, detailed process is as follows:
By calculating the edge point set sum S in target templateAAnd the marginal point of best match position region template
Gather sum SB, calculate Adaptive template-updating conditional parameter P1=SA/SBAnd the gray scale of target template region original image
Mean value hAWith the original image gray average h of best match position region templateB, calculate Adaptive template-updating conditional parameter P2
=hA/hB, and if only if meeting condition α < P1< β and condition γ < P2When < λ, update current region template is target template,
Otherwise it does not update, wherein α, beta, gamma, λ is constant.To reduce the interference of noise, real-time, stable infrared object tracking is carried out.
As shown in Fig. 2, in the method for the present invention, it is a kind of based on Hausdorff apart from matched adaptive template it is infrared with
Track method carries out the detection of objective contour using Sobel operator filtering template both horizontally and vertically and establishes contour mould,
Formula is as follows:
Horizontal gradient amplitude Sx=(Z1+2Z2+Z3)-(Z7+2Z8+Z9)
Vertical gradient amplitude Sy=(Z1+2Z4+Z7)-(Z3+2Z6+Z9)
Wherein, Z1,Z2,Z3,Z4,Z5,Z6,Z7,Z8,Z9The 8 neighborhood territory pixel gray values of respectively pixel f (i, j).
As shown in figure 3, in the method for the present invention, it is a kind of based on Hausdorff apart from matched adaptive template it is infrared with
Track method carries out the detection of objective contour using the Sobel operator filtering template in positive and negative 45 degree of directions and establishes contour mould, public
Formula is as follows:
Positive 45 degree of direction gradient amplitudes S45°=(Z2+2Z3+Z6)-(Z4+2Z7+Z8)
Minus 45 degree of direction gradient amplitudes S-45°=(Z2+2Z1+Z4)-(Z6+2Z8+Z9)
In short, in the present invention: the infrared track method based on Hausdorff apart from matched adaptive template, which uses, to be changed
Into Sobel edge detection template, enhance the ability of infrared target edge detection;Improved Hausdorff distance is used
Matching process greatly improves the operating rate of algorithm;Adaptive template-updating strategy increases the robustness of target following
By force, the precision of tracking is improved.
Claims (6)
1. a kind of infrared track method characterized by comprising
Step 1: input infrared video, the selected target region in the initial frame image of the infrared video, and to described initial
Frame image carries out binary conversion treatment, obtains the edge contour information of target and establishes target template;
Step 2: reading next frame image, binary conversion treatment is carried out to the next frame image, obtains the edge of area to be tested
Profile information;
Step 3: the target template and the area to be tested being matched, the position of best match is found out;
The two-way Hausdorff distance matching that the target template and the area to be tested are improved;
To the marginal information set B=of marginal information set A={ a1, a2 ..., am } and region to be measured in target template
{ b1, b2 ..., bn } carries out two-way Hausdorff distance matching, and formula is as follows:
H (A, B)=max (h (A, B), h (B, A))
Wherein, a1, a2 ..., am are the coordinate information of target template edge aggregation, and b1, b2 ..., bm are candidate template to be measured
Edge coordinate,
H (A, B)=max (a ∈ A) min (b ∈ B) | | a-b | |
H (B, A)=max (b ∈ B) min (a ∈ A) | | b-a | |
H (A, B) expression is arranged from the distance set of point ai to the point for putting nearest set B apart from this of set A in formula (6)
Sequence takes maximum value therein, in formula (7) h (B, A) indicate from the point bi of set B to the point for putting nearest set A apart from this away from
It is ranked up from set, takes maximum value therein, | | | | it is apart from normal form;
When the Hausdorff improved is apart from matching primitives, to being looked in another set B in 8 contiguous ranges of current point ai
Whether point positioned at this range is had, if four neighborhood positions up and down of ai, which correspond in set B, at least one corresponding point,
This ai point is calculated to set B point Hausdorff distance as 1, otherwise continually look for other upper lefts of ai eight neighborhood point, lower-left, upper right,
Whether corresponding at least one point is had in four neighbor assignment set B of bottom right, if so, then calculating this ai point to set B point
Hausdorff distance is 2, and otherwise distance is just denoted as 10;
Step 4: judging whether target template meets update condition using adaptive mode slat element, if meeting update condition more
Fresh target template simultaneously reads next frame image execution step 1 to step 4, does not otherwise update target template and reads next frame image
Step 1 is executed to step 4, until all images in the infrared video are all disposed.
2. the method as described in claim 1, which is characterized in that the step 1 includes: using improved Sobel operator to institute
It states initial frame image and carries out binary conversion treatment, set 1 for the edge pixel gray value of the target area, rest of pixels value is set
It is 0, obtains the edge contour information of target and establish target template.
3. method according to claim 2, which is characterized in that the step 1 includes:
For each pixel in the initial frame image, calculate separately:
Sx=(Z1+2Z2+Z3)-(Z7+2Z8+Z9)
Sy=(Z1+2Z4+Z7)-(Z3+2Z6+Z9)
S45 °=(Z2+2Z3+Z6)-(Z4+2Z7+Z8)
S-45 °=(Z2+2Z1+Z4)-(Z6+2Z8+Z9)
Wherein, Z1,Z2,Z3,Z4,Z5,Z6,Z7,Z8,Z9Respectively 8 neighborhood territory pixel gray values of current pixel point;
As the S of current pixel pointx,Sy,S45°,S-45°In any one be greater than preset threshold value when, current pixel point be edge picture
Element.
4. the method as described in claim 1, which is characterized in that the step 2 includes: using improved Sobel operator to institute
It states next frame image and carries out binary conversion treatment, set 1 for the edge pixel gray value in the next frame image, rest of pixels
Value is set as 0, obtains the edge contour information of area to be tested.
5. method as claimed in claim 4, which is characterized in that the step 2 includes:
For each pixel in the next frame image, calculate separately.
6. the method as described in claim 1, which is characterized in that step 4 includes:
By calculating the edge point set sum S in target templateAAnd the edge point set of best match position region template is total
Number SB, calculate Adaptive template-updating conditional parameter
P1=SA/SB, and pass through the gray average h of target template region original imageAWith best match position region template
Original image gray average hB, calculate Adaptive template-updating conditional parameter P2=hA/hB, and if only if meeting condition α < P1< β
With condition γ < P2When < λ, update current region template is target template, is not otherwise updated, wherein α, beta, gamma, λ is constant.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610803370.8A CN106447685B (en) | 2016-09-06 | 2016-09-06 | A kind of infrared track method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610803370.8A CN106447685B (en) | 2016-09-06 | 2016-09-06 | A kind of infrared track method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106447685A CN106447685A (en) | 2017-02-22 |
CN106447685B true CN106447685B (en) | 2019-04-02 |
Family
ID=58163947
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610803370.8A Expired - Fee Related CN106447685B (en) | 2016-09-06 | 2016-09-06 | A kind of infrared track method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106447685B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109741305B (en) * | 2018-12-26 | 2020-11-27 | 中国科学院宁波工业技术研究院慈溪生物医学工程研究所 | Method for detecting imaging damage image of capsule endoscope |
CN114566056B (en) * | 2022-02-28 | 2023-01-24 | 招商局重庆公路工程检测中心有限公司 | Highway tunnel driving safety risk identification, prevention and control method and system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1738426A (en) * | 2005-09-09 | 2006-02-22 | 南京大学 | Video motion goal division and track method |
CN101673403A (en) * | 2009-10-10 | 2010-03-17 | 安防制造(中国)有限公司 | Target following method in complex interference scene |
CN101739686A (en) * | 2009-02-11 | 2010-06-16 | 北京智安邦科技有限公司 | Moving object tracking method and system thereof |
CN102005054A (en) * | 2010-11-24 | 2011-04-06 | 中国电子科技集团公司第二十八研究所 | Real-time infrared image target tracking method |
CN103020956A (en) * | 2012-11-20 | 2013-04-03 | 华中科技大学 | Image matching method for judging Hausdorff distance based on decision |
CN104036524A (en) * | 2014-06-18 | 2014-09-10 | 哈尔滨工程大学 | Fast target tracking method with improved SIFT algorithm |
CN105184822A (en) * | 2015-09-29 | 2015-12-23 | 中国兵器工业计算机应用技术研究所 | Target tracking template updating method |
-
2016
- 2016-09-06 CN CN201610803370.8A patent/CN106447685B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1738426A (en) * | 2005-09-09 | 2006-02-22 | 南京大学 | Video motion goal division and track method |
CN101739686A (en) * | 2009-02-11 | 2010-06-16 | 北京智安邦科技有限公司 | Moving object tracking method and system thereof |
CN101673403A (en) * | 2009-10-10 | 2010-03-17 | 安防制造(中国)有限公司 | Target following method in complex interference scene |
CN102005054A (en) * | 2010-11-24 | 2011-04-06 | 中国电子科技集团公司第二十八研究所 | Real-time infrared image target tracking method |
CN103020956A (en) * | 2012-11-20 | 2013-04-03 | 华中科技大学 | Image matching method for judging Hausdorff distance based on decision |
CN104036524A (en) * | 2014-06-18 | 2014-09-10 | 哈尔滨工程大学 | Fast target tracking method with improved SIFT algorithm |
CN105184822A (en) * | 2015-09-29 | 2015-12-23 | 中国兵器工业计算机应用技术研究所 | Target tracking template updating method |
Non-Patent Citations (4)
Title |
---|
基于Hausdorff距离的目标跟踪算法与实现;尚俊;《湖北第二师范学院学报》;20130215;第30卷(第2期);46-48 |
基于机器视觉的印刷线路板缺陷检测技术研究;黄李;《中国优秀硕士论文全文数据库 信息科技辑》;20151015;正文第38页第2)部分,第41页第(2)部分 |
红外导引头边缘模板匹配跟踪算法研究;张凯 等;《西北工业大学学报》;20080615;第26卷(第3期);第331-333页摘要,第1-3节,图1 |
结合SURF与聚类分析方法实现运动目标的快速跟踪;李英 等;《液晶与显示》;20110815;第26卷(第4期);544-550 |
Also Published As
Publication number | Publication date |
---|---|
CN106447685A (en) | 2017-02-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110688987B (en) | Pedestrian position detection and tracking method and system | |
CN105678338B (en) | Target tracking method based on local feature learning | |
CN109657575B (en) | Intelligent video tracking algorithm for outdoor constructors | |
CN104200485B (en) | Video-monitoring-oriented human body tracking method | |
CN107369159B (en) | Threshold segmentation method based on multi-factor two-dimensional gray level histogram | |
CN106296725A (en) | Moving target detects and tracking and object detecting device in real time | |
CN104966304B (en) | Multi-target detection tracking based on Kalman filtering and nonparametric background model | |
CN109086724B (en) | Accelerated human face detection method and storage medium | |
CN110610150B (en) | Tracking method, device, computing equipment and medium of target moving object | |
KR100651034B1 (en) | System for detecting targets and method thereof | |
CN106204633B (en) | Student tracking method and device based on computer vision | |
CN103456009B (en) | Object detection method and device, supervisory system | |
CN103985142A (en) | Federated data association Mean Shift multi-target tracking method | |
CN106023248A (en) | Real-time video tracking method | |
CN108133491A (en) | A kind of method for realizing dynamic target tracking | |
CN108364300A (en) | Vegetables leaf portion disease geo-radar image dividing method, system and computer readable storage medium | |
CN106447685B (en) | A kind of infrared track method | |
CN113608663A (en) | Fingertip tracking method based on deep learning and K-curvature method | |
CN103577833A (en) | Abnormal intrusion detection method based on motion template | |
CN109658441B (en) | Foreground detection method and device based on depth information | |
Koniar et al. | Machine vision application in animal trajectory tracking | |
US20230419500A1 (en) | Information processing device and information processing method | |
CN110688969A (en) | Video frame human behavior identification method | |
CN103578121A (en) | Motion detection method based on shared Gaussian model in disturbed motion environment | |
KR101542206B1 (en) | Method and system for tracking with extraction object using coarse to fine techniques |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
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: 20190402 |