CN107169990A - A kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm - Google Patents
A kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm Download PDFInfo
- Publication number
- CN107169990A CN107169990A CN201710266254.1A CN201710266254A CN107169990A CN 107169990 A CN107169990 A CN 107169990A CN 201710266254 A CN201710266254 A CN 201710266254A CN 107169990 A CN107169990 A CN 107169990A
- Authority
- CN
- China
- Prior art keywords
- particle
- spot
- mrow
- pixel
- optimization algorithm
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
- G06T7/251—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of video multiple mobile object localization method based on particle swarm optimization algorithm.This method carries out figure spot detection by Background difference first, then by the way that population to be divided into particle subgroup in units of figure spot, and particle swarm optimization algorithm is used in each subgroup, to reach the purpose of multiple mobile object positioning.The inventive method can solve the problem that multiple target acquisition problems and because object is deadlocked, picture it is dim caused by detection difficult problem.
Description
Technical field
The present invention relates to technical field of image processing, particularly a kind of video based on particle swarm optimization algorithm is done more physical exercises mesh
Mark method for tracking and positioning.
Background technology
Particle swarm optimization algorithm (PSO) is inspired by the regularity of flying bird cluster activity, and then is built using swarm intelligence
A vertical simplified model.Particle cluster algorithm is on the basis of to animal cluster activity behavior observation, using individual right in colony
The shared motion for making whole colony of information produces the evolutionary process from disorder to order in problem solving space, so as to obtain
Optimal solution.PSO is similar with genetic algorithm, is a kind of optimized algorithm based on iteration.System initialization is one group of RANDOM SOLUTION, is passed through
Iterated search optimal value.But it does not have the intersection of genetic algorithm and variation, but particle follows optimal in solution space
Particle is scanned for.Compare with genetic algorithm, PSO advantage is simple easily realization and needs to adjust without many parameters
It is whole.Function optimization is widely used at present, and neural metwork training, fuzzy system control and the application of other genetic algorithms are led
Domain.
However, traditional particle cluster algorithm can only solve single goal orientation problem, and Multi-target position can not be solved the problems, such as.
The content of the invention
The technical problems to be solved by the invention are to overcome the deficiencies in the prior art and provide a kind of based on particle group optimizing
The video multiple mobile object method for tracking and positioning of algorithm, combines particle swarm optimization algorithm using figure spot detection, population is split
The purpose of Multi-target position is reached into the subgroup of respective numbers.
The present invention uses following technical scheme to solve above-mentioned technical problem：
According to a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm proposed by the present invention, bag
Include following steps：
Step 1, user's input video, the f two field pictures in video are uniformly extracted by same time interval, and access by row is each
Each pixel of two field picture, records the colouring intensity value of three passages of each pixel, calculates the gray scale of each pixel
Change value, makes each two field picture be converted into a two-dimensional matrix, the i-th two field picture is designated as into Ii, i ∈ { 1,2 ..., f }；
Step 2, will be divided into figure spot pixel and background pixel point per two field picture successively, obtain figure spot quantity B and its
Geometric properties；The figure spot pixel is pixel where moving target, and background pixel point is pixel where non-athletic target；
It is specific as follows：
Step 2.1, it regard the first two field picture as background image B0(x, y), sets threshold value T；
Step 2.2, successively by i value from 1 to f-1, using formulaObtain IiWith Ii+1Between frame it is poor
Bianry image Di, by the poor bianry image D of frameiUpdate the background image B of the i-th framei(x, y),α
For renewal speed；
Step 2.3, by Bf-1(x, y) is considered as background image B (x, y)；
Step 2.4, successively by i values 1 arrive f-1, using formulaMeter
Calculate background difference bianry image DBi(x,y)；In background difference bianry image DBiIn (x, y), each pixel is scanned, by DBi
All gray values are referred to as background pixel point for 0 pixel in (x, y), by DBiAll gray values are 255 picture in (x, y)
Vegetarian refreshments is referred to as figure spot pixel, and the connected domain that figure spot pixel is constituted is referred to as figure spot, obtains the quantity B of figure spot, while obtaining every
The geometric properties of individual figure spot, the geometric properties include line segment boundary point, boundary rectangle, area and the position of form center of figure spot；
Step 3, setting initialization iterations k are 1；
Step 4, detection kth frame input picture Ik, carry out the initialization of particle：It is random in the boundary rectangle of each figure spot
The population in uniform particle, the boundary rectangle of m-th of figure spot is generated for pop (m), pop (m)=min ((axis (m)/
axismin)*popmin,popmax), wherein, axis (m) is the length of the boundary rectangle of figure spot, axisminIt is the external of all figure spots
During long minimum long in of rectangle, popminAnd popmaxIt is the minimum and most population of each population in allowed band,
Total number of particlesIf k=1, the position P of each particle of random initializtion0With speed V0If, k ＞ 1, respectively
Particle is initialized as the position P of all particles preserved during the frame of kth -1k-1With speed Vk-1；
Step 5, with particle swarm optimization algorithm, carry out M iteration；It is specific as follows：
Step 5.1, initialization iterations t are 1；
Step 5.2, the fitness for calculating particle
Wherein Represent the t times
J-th of particle during iteration, λ is weight coefficient,It is the cost of the color histogram of figure spot where j-th of particle,It is the gradient cost of the gradient direction of figure spot where j-th of particle, j=1,2,3 ... Q；
Step 5.3, successively by j value from 1 to Q, update the local optimum solution of each particle：As t=1,
The local optimum scheme of the t times iterationAs t ＞ 1, the local optimum side of the t times iteration of j-th of particle
Case is
Update the overall preferred plan of the t times iteration：During t=1, gbest (t)=Pt, wherein, P1The position of all particles during for initialization
Put matrix；As t ＞ 1, the overall preferred plan of the t times iteration is
Step 5.4, pass through calculateTo update all grains
The position P of the t times of sontWith speed Vt；Wherein, w is weight coefficient, c1And c2It is acceleration constant, r1And r2It is by random value 1
Or 0 composition matrix,The local optimum scheme reached for all particles of the t-1 times iteration,When being the t-1 times iteration
Overall preferred plan, VtAnd Vt-1The speed of all particles when being the t times iteration and the t-1 times iteration；
Step 5.5, when iterations t is not reaching to default maximum iteration M, t=t+1, return to step 5.3；If t
=M, then perform step 5.6；
Step 5.6, preservation kth frame input picture IkIn each particle position PMWith speed VM, and preserve kth frame input
Image IkIn the boundary rectangle of figure spot where each particle four apex coordinate values；
Step 5.7, as k ＜ f, k from increase 1, return to step 4, otherwise terminate detection.
Enter as a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm of the present invention
One-step optimization scheme, r1And r2The matrix of the Q being made up of random value 1 or 0 × 4.
Enter as a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm of the present invention
One-step optimization scheme,For the matrix of Q × 4.
Enter as a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm of the present invention
One-step optimization scheme,It is the matrix of Q × 4.
Enter as a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm of the present invention
One-step optimization scheme, in the matrix of Q × 4, each particle of correspondence per a line, each dimension of one particle of each row correspondence.
The present invention uses above technical scheme compared with prior art, with following technique effect：
(1) the multiple mobile object localization method of the present invention based on particle swarm optimization algorithm, solves traditional grain
The problem of subgroup optimized algorithm is unable to position portion moving target.Particle swarm optimization algorithm proposed by the present invention has crossed repetition erasing
The process of the object detected, realizes position portion and is blocked target, the deadlocked problem that solves multiple objects；
(2) present invention is using the population detector guided based on figure spot feature, due to all characteristic patterns normalizing
Change to less space scale, it is meant that low spatial resolution, therefore the pixel for needing to scan can be reduced, so as to effectively reduce
The calculating cost of algorithm.
Brief description of the drawings
Fig. 1 is the flow chart of the video multiple mobile object localization method based on particle swarm optimization algorithm.
Embodiment
Technical scheme is described in further detail below in conjunction with the accompanying drawings：
It is the flow chart of the video multiple mobile object localization method based on particle swarm optimization algorithm as shown in Figure 1, specifically such as
Under：
Step 1), from vehicle monitoring video recording in choose one section of 8046 frame video sequence, compression sizes to 240x320 pixels,
40 frame pictures in video are uniformly extracted by same time interval.Each pixel of each frame picture of access by row, record
The colouring intensity value of three passages of each pixel, calculates the gray processing value of each pixel, is converted into each two field picture
One two-dimensional matrix, the two-dimensional matrix that the i-th two field picture is changed into is designated as Ii, i ∈ { 1,2 ..., 40 }；
Step 2), will be divided into figure spot pixel and background pixel point per two field picture successively, obtain figure spot quantity and its
Geometric properties, the figure spot pixel is pixel where moving target, and background pixel point is pixel where non-athletic target；
Step 2) comprise the following steps that：
Step 2.1), regard the first two field picture as background image B0(x, y), sets threshold value T=27；
Step 2.2), successively by i values 1 arrive f-1, using formulaObtain IiWith Ii+1Between
The poor bianry image D of framei, by the poor bianry image D of frameiUpdate background image
Renewal speed α=0.6 is set.
Step 2.3), B39(x, y) can be considered background image matrix B (x, y)；
Step 2.4), successively by i values 1 to 39, using formulaMeter
Calculate background difference bianry image DBi(x,y)；In background difference bianry image DBiIn (x, y), each pixel is scanned, by DBi
All gray values are referred to as background pixel point for 0 pixel in (x, y), by DBiAll gray values are 255 picture in (x, y)
Vegetarian refreshments is referred to as figure spot pixel, and the connected domain that figure spot pixel is constituted is referred to as figure spot, and the quantity for obtaining figure spot is 5, is obtained simultaneously
The geometric properties of each figure spot, the geometric properties include line segment boundary point, boundary rectangle, area and the position of form center of figure spot；
Step 3), set iterations k, k is initialized as 1；
Step 4), detection kth frame input picture Ik, carry out the initialization of particle：In the boundary rectangle of each figure spot with
Machine generates the population in uniform particle, the boundary rectangle of m-th of figure spot for pop (m), pop (m)=min ((axis (m)/
axismin)*popmin,popmax), it is about 8, wherein axis to calculate the population in each rectanglemin=80, particle is total
Number Q=40；If k=1, the position P of each particle of random initializtion0With speed V0If k ＞ 1, each particle is initialized as
The position P of all particles preserved during k-1 framesk-1With speed Vk-1；
Step 5), with particle swarm optimization algorithm, carry out 16 iteration, flow is as follows：
Step 5.1, initialization iterations t are 1；
Step 5.2,Wherein
J-th of particle during the t times iteration is represented, λ is weight coefficient,Be figure spot where particle color histogram into
This,It is the gradient cost of the gradient direction of figure spot where particle；
Step 5.3, successively by j values 1 to 40, update the local optimum solution of each particle：As t=1,As t ＞ 1, zbest1The local optimum scheme of the t times iteration of j-th of particle is
Update the overall preferred plan of the t times iteration：During t=1, gbest (t)=Pt, wherein P1All grains during for initialization
The location matrix of son, P1For；As t ＞ 1, the overall preferred plan of the t times iteration is
When iterations j is not reaching to number of particles 8, j=j+1, step 5.2 is returned to；If t=8, step 5.3 is performed；
Step 5.4, pass through calculateTo update each grain
The state and speed of son.Set acceleration constant c1=1, c2=5, weight coefficient w=1, be made up of random value 1 or 08 × 4 square
Battle array r1And r2。
Step 5.5, when iterations t is not reaching to default maximum iteration 16, t=t+1, return to step 5.3；If
T=16, then perform step 5.6；
Step 5.6, preservation kth frame input picture IkIn each particle position PMWith speed VM, and preserve kth frame input
Image IkIn the boundary rectangle of figure spot where each particle four apex coordinate values；
Step 5.7, as k ＜ f, k from increase 1, return to step 4, otherwise terminate detection.The last frame image finally obtained
In 5 sports cars boundary rectangle apex coordinate value be { (2,12), (2,60), (24,12), (24,60) }, (12,
23), (12,69), (27,23), (27,69) }, { (25,78), (25,141), (47,78), (24,141) }, (121,35),
(121,65), (134,35), (134,65) }, { (205,267), (205,298), (234,267), (234,298) }.
Above content is to combine specific preferred embodiment further description made for the present invention, it is impossible to assert
The specific implementation of the present invention is confined to these explanations.For general technical staff of the technical field of the invention,
On the premise of not departing from present inventive concept, some simple deductions or replacement can also be made, should all be considered as belonging to the present invention's
Protection domain.
Claims (5)
1. a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm, it is characterised in that including as follows
Step：
Step 1, user's input video, the f two field pictures in video, each frame figure of access by row are uniformly extracted by same time interval
Each pixel of picture, records the colouring intensity value of three passages of each pixel, calculates the gray processing value of each pixel,
Each two field picture is converted into a two-dimensional matrix, the i-th two field picture is designated as Ii, i ∈ { 1,2 ..., f }；
Step 2, figure spot pixel and background pixel point will be divided into per two field picture, obtain the quantity B and its geometry of figure spot successively
Feature；The figure spot pixel is pixel where moving target, and background pixel point is pixel where non-athletic target；Specifically
It is as follows：
Step 2.1, it regard the first two field picture as background image B0(x, y), sets threshold value T；
Step 2.2, successively by i value from 1 to f-1, using formulaObtain IiWith Ii+1Between frame poor two
It is worth image Di, by the poor bianry image D of frameiUpdate the background image B of the i-th framei(x, y),α
For renewal speed；
Step 2.3, by Bf-1(x, y) is considered as background image B (x, y)；
Step 2.4, successively by i values 1 arrive f-1, using formulaCalculate the back of the body
Scape difference bianry image DBi(x,y)；In background difference bianry image DBiIn (x, y), each pixel is scanned, by DBi(x,y)
In all gray values for 0 pixel be referred to as background pixel point, by DBiAll gray values claim for 255 pixel in (x, y)
For figure spot pixel, the connected domain that figure spot pixel is constituted is referred to as figure spot, the quantity B of figure spot is obtained, while obtaining each figure spot
Geometric properties, the geometric properties include line segment boundary point, boundary rectangle, area and the position of form center of figure spot；
Step 3, setting initialization iterations k are 1；
Step 4, detection kth frame input picture Ik, carry out the initialization of particle：Generated at random in the boundary rectangle of each figure spot
Population in uniform particle, the boundary rectangle of m-th of figure spot is pop (m), pop (m)=min ((axis (m)/axismin)*
popmin,popmax), wherein, axis (m) is the length of the boundary rectangle of figure spot, axisminIt is the length of the boundary rectangle of all figure spots
During minimum long in, popminAnd popmaxIt is the minimum and most population of each population, total number of particles in allowed bandIf k=1, the position P of each particle of random initializtion0With speed V0If k ＞ 1, each particle is initial
Turn to the position P of all particles preserved during the frame of kth -1k-1With speed Vk-1；
Step 5, with particle swarm optimization algorithm, carry out M iteration；It is specific as follows：
Step 5.1, initialization iterations t are 1；
Step 5.2, the fitness for calculating particle
Wherein Represent the t times repeatedly
For when j-th of particle, λ is weight coefficient,It is the cost of the color histogram of figure spot where j-th of particle,It is the gradient cost of the gradient direction of figure spot where j-th of particle, j=1,2,3 ... Q；
Step 5.3, successively by j value from 1 to Q, update the local optimum solution of each particle：As t=1, the t times
The local optimum scheme of iterationAs t ＞ 1, the local optimum scheme of the t times iteration of j-th of particle is
Update the overall preferred plan of the t times iteration：During t=1, gbest (t)=Pt, wherein, P1All particles during for initialization
Location matrix；As t ＞ 1, the overall preferred plan of the t times iteration is
<mrow>
<mi>g</mi>
<mi>b</mi>
<mi>e</mi>
<mi>s</mi>
<mi>t</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>min</mi>
<mo>{</mo>
<mi>f</mi>
<mi>i</mi>
<mi>t</mi>
<mi>n</mi>
<mi>e</mi>
<mi>s</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<msubsup>
<mi>P</mi>
<mn>1</mn>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msubsup>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>f</mi>
<mi>i</mi>
<mi>t</mi>
<mi>n</mi>
<mi>e</mi>
<mi>s</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<msubsup>
<mi>P</mi>
<mn>2</mn>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msubsup>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mn>...</mn>
<mo>,</mo>
<mi>f</mi>
<mi>i</mi>
<mi>t</mi>
<mi>n</mi>
<mi>e</mi>
<mi>s</mi>
<mi>s</mi>
<mrow>
<mo>(</mo>
<msubsup>
<mi>P</mi>
<mi>Q</mi>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msubsup>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>g</mi>
<mi>b</mi>
<mi>e</mi>
<mi>s</mi>
<mi>t</mi>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>}</mo>
<mo>;</mo>
</mrow>
Step 5.4, pass through calculateTo update all particles
The position P of the t timestWith speed Vt；Wherein, w is weight coefficient, c1And c2It is acceleration constant, r1And r2It is by random value 1 or 0
The matrix of composition,The local optimum scheme reached for all particles of the t-1 times iteration,It is whole when being the t-1 times iteration
Body preferred plan, VtAnd Vt-1The speed of all particles when being the t times iteration and the t-1 times iteration；
Step 5.5, when iterations t is not reaching to default maximum iteration M, t=t+1, return to step 5.3；If t=M,
Then perform step 5.6；
Step 5.6, preservation kth frame input picture IkIn each particle position PMWith speed VM, and preserve kth frame input picture Ik
In the boundary rectangle of figure spot where each particle four apex coordinate values；
Step 5.7, as k ＜ f, k from increase 1, return to step 4, otherwise terminate detection.
2. a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm according to claim 1,
Characterized in that, r1And r2The matrix of the Q being made up of random value 1 or 0 × 4.
3. a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm according to claim 1,
Characterized in that,For the matrix of Q × 4.
4. a kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm according to claim 1,
Characterized in that,It is the matrix of Q × 4.
5. a kind of video multiple mobile object track and localization based on particle swarm optimization algorithm according to Claims 2 or 3 or 4
Method, it is characterised in that in the matrix of Q × 4, each particle of correspondence per a line, each dimension of one particle of each row correspondence
Degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710266254.1A CN107169990A (en) | 2017-04-21 | 2017-04-21 | A kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710266254.1A CN107169990A (en) | 2017-04-21 | 2017-04-21 | A kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107169990A true CN107169990A (en) | 2017-09-15 |
Family
ID=59813316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710266254.1A Pending CN107169990A (en) | 2017-04-21 | 2017-04-21 | A kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107169990A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107742306A (en) * | 2017-09-20 | 2018-02-27 | 徐州工程学院 | Moving Target Tracking Algorithm in a kind of intelligent vision |
| CN109146922A (en) * | 2018-07-11 | 2019-01-04 | 哈尔滨工程大学 | A kind of Forward-looking Sonar Method for Underwater Target Tracking based on adaptive particle swarm optimization |
| CN117152258A (en) * | 2023-11-01 | 2023-12-01 | 中国电建集团山东电力管道工程有限公司 | Product positioning method and system for intelligent workshop of pipeline production |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663775A (en) * | 2012-03-30 | 2012-09-12 | 温州大学 | Target tracking method oriented to video with low frame rate |
| US8620631B2 (en) * | 2011-04-11 | 2013-12-31 | King Fahd University Of Petroleum And Minerals | Method of identifying Hammerstein models with known nonlinearity structures using particle swarm optimization |
| US20140257767A1 (en) * | 2013-03-09 | 2014-09-11 | Bigwood Technology, Inc. | PSO-Guided Trust-Tech Methods for Global Unconstrained Optimization |
| CN105069441A (en) * | 2015-08-21 | 2015-11-18 | 燕山大学 | Moving vehicle detection method based on background updating and particle swarm optimization algorithm |
-
2017
- 2017-04-21 CN CN201710266254.1A patent/CN107169990A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8620631B2 (en) * | 2011-04-11 | 2013-12-31 | King Fahd University Of Petroleum And Minerals | Method of identifying Hammerstein models with known nonlinearity structures using particle swarm optimization |
| CN102663775A (en) * | 2012-03-30 | 2012-09-12 | 温州大学 | Target tracking method oriented to video with low frame rate |
| US20140257767A1 (en) * | 2013-03-09 | 2014-09-11 | Bigwood Technology, Inc. | PSO-Guided Trust-Tech Methods for Global Unconstrained Optimization |
| CN105069441A (en) * | 2015-08-21 | 2015-11-18 | 燕山大学 | Moving vehicle detection method based on background updating and particle swarm optimization algorithm |
Non-Patent Citations (2)
| Title |
|---|
| MYO THIDA ET AL.: "A Particle Swarm Optimization Approach for Multi-objects Tracking in Crowded Scene", 《2009 IEEE 12TH INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS》 * |
| 张顺淼 等: "一种基于Surendra背景更新的背景减除运动目标检测方法", 《南京工程学院学报（自然科学版）》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107742306A (en) * | 2017-09-20 | 2018-02-27 | 徐州工程学院 | Moving Target Tracking Algorithm in a kind of intelligent vision |
| CN107742306B (en) * | 2017-09-20 | 2021-03-23 | 徐州工程学院 | Moving target tracking algorithm in intelligent vision |
| CN109146922A (en) * | 2018-07-11 | 2019-01-04 | 哈尔滨工程大学 | A kind of Forward-looking Sonar Method for Underwater Target Tracking based on adaptive particle swarm optimization |
| CN109146922B (en) * | 2018-07-11 | 2021-07-06 | 哈尔滨工程大学 | Forward-looking sonar underwater target tracking method based on adaptive particle swarm optimization |
| CN117152258A (en) * | 2023-11-01 | 2023-12-01 | 中国电建集团山东电力管道工程有限公司 | Product positioning method and system for intelligent workshop of pipeline production |
| CN117152258B (en) * | 2023-11-01 | 2024-01-30 | 中国电建集团山东电力管道工程有限公司 | Product positioning method and system for intelligent workshop of pipeline production |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104899255B (en) | Suitable for the construction method of the image data base of training depth convolutional neural networks | |
| Al Bashish et al. | A framework for detection and classification of plant leaf and stem diseases | |
| CN103942557B (en) | A kind of underground coal mine image pre-processing method | |
| Horng et al. | Multilevel image thresholding selection based on the firefly algorithm | |
| Kandhway et al. | A water cycle algorithm-based multilevel thresholding system for color image segmentation using masi entropy | |
| CN107169990A (en) | A kind of video multiple mobile object method for tracking and positioning based on particle swarm optimization algorithm | |
| CN110517311A (en) | Pest and disease monitoring method based on leaf spot lesion area | |
| CN110427922A (en) | One kind is based on machine vision and convolutional neural networks pest and disease damage identifying system and method | |
| ya Guo et al. | Two-dimensional Otsu's thresholding segmentation method based on grid box filter | |
| CN107590442A (en) | A kind of video semanteme Scene Segmentation based on convolutional neural networks | |
| CN110796678B (en) | Underwater multi-target tracking method based on IoU | |
| CN109410168A (en) | For determining the modeling method of the convolutional neural networks model of the classification of the subgraph block in image | |
| CN104143186B (en) | A kind of SLIC super-pixel piecemeal optimization method | |
| CN113222991A (en) | Deep learning network-based field ear counting and wheat yield prediction | |
| CN105631899B (en) | A kind of ultrasound image motion target tracking method based on gray scale textural characteristics | |
| CN107491734A (en) | Semi-supervised Classification of Polarimetric SAR Image method based on multi-core integration Yu space W ishart LapSVM | |
| CN113344972B (en) | Fish track prediction method based on intensive culture | |
| CN113822198B (en) | Peanut growth monitoring method, system and medium based on UAV-RGB image and deep learning | |
| CN107351080A (en) | A kind of hybrid intelligent research system and control method based on array of camera units | |
| CN112580662A (en) | Method and system for recognizing fish body direction based on image features | |
| CN112469050A (en) | WSN three-dimensional coverage enhancement method based on improved wolf optimizer | |
| Roggiolani et al. | Hierarchical approach for joint semantic, plant instance, and leaf instance segmentation in the agricultural domain | |
| Yang | RETRACTED ARTICLE: Research on multi-target tracking technology based on machine vision | |
| CN110287990A (en) | Microalgae image classification method, system, equipment and storage medium | |
| Ban et al. | A lightweight model based on YOLOv8n in wheat spike detection |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170915 |
|
| WD01 | Invention patent application deemed withdrawn after publication |