CN110543015A - Non-three-dimensional imaging binocular vision infrared image target tracking device and method - Google Patents
Non-three-dimensional imaging binocular vision infrared image target tracking device and method Download PDFInfo
- Publication number
- CN110543015A CN110543015A CN201910845157.7A CN201910845157A CN110543015A CN 110543015 A CN110543015 A CN 110543015A CN 201910845157 A CN201910845157 A CN 201910845157A CN 110543015 A CN110543015 A CN 110543015A
- Authority
- CN
- China
- Prior art keywords
- target
- image sensor
- image
- time
- target tracking
- 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
- 238000003384 imaging method Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000033001 locomotion Effects 0.000 claims abstract description 13
- 230000005540 biological transmission Effects 0.000 claims description 2
- 230000003287 optical effect Effects 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000002349 favourable effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B27/00—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00
- G02B27/0093—Optical systems or apparatus not provided for by any of the groups G02B1/00 - G02B26/00, G02B30/00 with means for monitoring data relating to the user, e.g. head-tracking, eye-tracking
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Closed-Circuit Television Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
the invention discloses a non-three-dimensional imaging binocular vision infrared image target tracking device and method, and belongs to the technical field of image processing and target tracking. The device comprises a collimating lens, a half-reflecting and half-transmitting lens, a first imaging objective lens and a first image sensor, wherein the collimating lens and the half-reflecting and half-transmitting lens are sequentially arranged along the light propagation direction; the imaging system also comprises a reflecting prism, a second imaging objective lens and a second image sensor which are arranged on the semi-reflecting and semi-transmitting mirror reflection optical path, wherein the image plane of the second image sensor is superposed with the focal plane of the second imaging objective lens; the first image sensor is rotatable; the method comprises the steps that firstly, whether a target rotates or not is judged by using a second image sensor, and whether the target moves along or in the direction vertical to the radius or not is judged by using two parameters obtained by a first image sensor under the condition of no rotation; the invention can identify the two-dimensional motion direction of the non-rotating target and is used for realizing target tracking.
Description
Technical Field
A non-three-dimensional imaging binocular vision infrared image target tracking device and method belong to the technical field of image processing and target tracking.
background
visual target tracking is an important research direction in computer vision, and has wide application in the fields of video monitoring, man-machine interaction, unmanned driving and the like.
in the last two thirty years, the machine vision target tracking technology has made great progress, and particularly in the last two years, a satisfactory effect is achieved by using a deep learning target tracking method, so that the target tracking technology has made a breakthrough progress.
however, moving object tracking is a very challenging task because the moving scene is very complex and changes frequently for a moving object, or the object itself changes continuously. In addition, factors such as illumination, low resolution, motion blur, beyond-field, rotation and the like also bring difficulty to target tracking.
for tracking military targets, such as tracking target vehicles on open land and tracking ships on the sea, the method has important strategic significance, and often utilizes the characteristic that the temperature of the target is different from the background, and infrared images are often adopted for tracking the targets, so that influence factors such as shielding, background mottle and the like can be effectively avoided.
however, there is still two-dimensional motion and rotation of the target in the horizontal plane. Because only the coordinates of a military target are concerned, how to acquire the two-position motion of the target under the influence of rotation is a technical problem to be solved urgently in the field.
disclosure of Invention
Aiming at the technical requirements, the invention discloses a non-three-dimensional imaging binocular vision infrared image target tracking device and method, and also discloses a binocular vision bracket and a target turning judgment method, wherein when a target has two-dimensional motion and turning in a horizontal plane, the two-dimensional motion direction of the target is judged.
The purpose of the invention is realized as follows:
a non-three-dimensional imaging binocular vision infrared image target tracking device comprises a collimating lens, a half-reflecting half-transmitting lens, a first imaging objective lens and a first image sensor, wherein the collimating lens and the half-reflecting half-transmitting lens are sequentially arranged along the light propagation direction; the optical lens system further comprises a reflecting prism, a second imaging objective lens and a second image sensor which are arranged on the semi-reflecting and semi-transmitting mirror reflection optical path, wherein the image plane of the second image sensor is superposed with the focal plane of the second imaging objective lens; the first image sensor can rotate at a constant speed by taking a vertical direction as an axis.
a binocular vision support comprises an upper plate and a lower plate which are sequentially and horizontally arranged from top to bottom, wherein the two sides of the upper plate and the lower plate are connected through a connecting rod, a second image sensor is fixedly installed at the top of the upper plate, a through hole is formed in the lower plate, a bearing which is installed in an interference fit mode with the through hole is arranged in the through hole, a rotating shaft is installed in the bearing in an interference fit mode, the lower portion of the rotating shaft is connected with a motor through a coupler, and a first image sensor is installed above the rotating shaft through threads; the motor rotates at a constant speed to drive the rotating shaft to rotate at a constant speed, and further the first image sensor is driven to rotate at a constant speed by taking the vertical direction as an axis.
A non-three-dimensional imaging binocular vision infrared image target tracking method requires a first image sensor to image under the condition of uniform rotation by taking a vertical direction as an axis, and the rotation angular speed is omega; the second image sensor is used for imaging under a fixed condition; the method comprises the following steps:
Step a, judging whether a target has a rotating body by using a second image sensor, if:
If yes, returning to the step a;
If not, entering the step b;
b, recording the time tn,1 when the target enters the field of view of the first image sensor for the nth time and the time tn,2 when the target leaves the field of view of the first image sensor for the nth time; recording the time tn +1,1 when the target enters the field of view of the first image sensor at the n +1 th time and the time tn +1,2 when the target leaves the field of view of the first image sensor at the n +1 th time;
Step c, judging the size between tn,2-tn,1 and tn +1,2-tn +1, if:
tn,2-tn,1 > tn +1,2-tn +1,1, the target is far away from the target tracking device along the radius direction;
tn,2-tn,1 > tn +1,2-tn +1,1, the target being close to the target tracking means in the radial direction;
tn,2-tn,1 > tn +1,2-tn +1,1, no motion of the target along the radius direction;
The radius direction is the connecting direction of the target and the target tracking device;
d, judging the size between the step d and the step d, if:
The target moves along the rotation direction of the first image sensor in the vertical radius direction;
the target moves along the direction vertical to the radius against the rotation direction of the first image sensor;
no movement is generated in the direction of vertical radius of the target;
The radius direction is a connecting direction of the target and the target tracking device.
An image-based target swivel judgment method comprises the following steps:
step a, when a target enters a first image sensor for the nth time, a second image sensor images to obtain an image f1, and when the target enters the first image sensor for the (n + 1) th time, the second image sensor images to obtain an image f 2;
b, performing 2 × 2 expansion on the image f1 to obtain an image f 3;
Step c, calculating according to the following formula
where M N is the resolution of f1 or image f 2; i and j represent the ith row and jth column of the image, respectively; x is 1,2, …, M, y is 1,2, …, N;
step d, judging whether max [ r (x, y) ] is smaller than a threshold value, if:
it is, the target does not have a swivel;
otherwise, the target has a swivel.
Has the advantages that:
firstly, the binocular vision technology is adopted, but the binocular vision technology is different from the traditional binocular vision technology in that the binocular vision of the invention is operated independently, and each camera is responsible for processing respective information, so that the arrangement simplifies the work of calibration and the like of the binocular vision;
secondly, the method can take the target entering the view field and leaving the view field as favorable factors on the basis of not establishing a three-dimensional image, can judge the two-dimensional movement directions of the target by only acquiring four time parameters, has simple calculation and high operation speed, and is favorable for tracking the high-speed moving target;
thirdly, whether the target has a rotating body can be judged only through a core formula, so that the influence on the judgment of the target position is avoided when the target translates and rotates simultaneously;
The invention also provides a structure of the binocular vision bracket.
drawings
FIG. 1 is a light path diagram of a non-three-dimensional imaging binocular vision infrared image target tracking device of the invention.
fig. 2 is a schematic structural view of the binocular vision bracket of the present invention.
In the figure: the device comprises a collimator 1, a half-reflecting half-transmitting mirror 2, a first imaging objective 3, a first image sensor 4, a reflecting prism 5, a second imaging objective 6, a second image sensor 7, an upper plate 11, a lower plate 12, a connecting rod 13, a bearing 14, a rotating shaft 15 and a motor 16.
Detailed Description
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
detailed description of the invention
the embodiment is an example of a non-three-dimensional imaging binocular vision infrared image target tracking device.
the optical path diagram of the non-three-dimensional imaging binocular vision infrared image target tracking device of the embodiment is shown in fig. 1. The device comprises a collimator lens 1 and a semi-reflecting and semi-transmitting lens 2 which are sequentially arranged along the light propagation direction, and further comprises a first imaging objective lens 3 and a first image sensor 4 which are arranged on the transmission light path of the semi-reflecting and semi-transmitting lens 2, wherein the image plane of the first image sensor 4 is superposed with the focal plane of the first imaging objective lens 3; the imaging device further comprises a reflecting prism 5, a second imaging objective 6 and a second image sensor 7 which are arranged on a reflection light path of the half-reflecting and half-transmitting mirror 2, wherein an image plane of the second image sensor 7 is superposed with a focal plane of the second imaging objective 6; the first image sensor 4 can rotate at a constant speed with a vertical direction as an axis.
detailed description of the invention
the present embodiment is an example of a binocular vision support.
the binocular vision support of the embodiment is shown in fig. 2 in a schematic structural diagram, and comprises an upper plate 11 and a lower plate 12 which are sequentially horizontally arranged from top to bottom, two sides of the upper plate 11 and the lower plate 12 are connected through a connecting rod 13, a second image sensor 7 is fixedly mounted at the top of the upper plate 11, a through hole is formed in the lower plate 12, a bearing 14 which is in interference fit with the through hole is arranged in the through hole, a rotating shaft 15 is mounted in the bearing 14 in interference fit with an inner ring, the lower part of the rotating shaft 15 is connected with a motor 16 through a coupler, and a first image sensor 4 is mounted above the rotating shaft 15 through threads; the motor 16 rotates at a constant speed to drive the rotating shaft 15 to rotate at a constant speed, and further drive the first image sensor 4 to rotate at a constant speed by taking the vertical direction as an axis.
It should be noted that the binocular vision bracket of the present embodiment may be implemented alone, or may be implemented on the basis of the first specific embodiment.
detailed description of the invention
The embodiment is an embodiment of a non-three-dimensional imaging binocular vision infrared image target tracking method.
the non-three-dimensional imaging binocular vision infrared image target tracking method of the embodiment requires that the first image sensor 4 images under the condition of uniform rotation by taking the vertical direction as an axis, and the rotation angular speed is omega; the second image sensor 7 images under stationary conditions; the method comprises the following steps:
Step a, judging whether a target has a rotating body by using a second image sensor 7, if:
If yes, returning to the step a;
If not, entering the step b;
B, recording the time tn,1 when the target enters the field of view of the first image sensor 4 for the nth time and the time tn,2 when the target leaves the field of view of the first image sensor 4 for the nth time; recording the time tn +1,1 when the target enters the field of view of the first image sensor 4 at the n +1 th time and the time tn +1,2 when the target leaves the field of view of the first image sensor 4 at the n +1 th time;
Step c, judging the size between tn,2-tn,1 and tn +1,2-tn +1, if:
tn,2-tn,1 > tn +1,2-tn +1,1, the target is far away from the target tracking device along the radius direction;
tn,2-tn,1 > tn +1,2-tn +1,1, the target being close to the target tracking means in the radial direction;
tn,2-tn,1 > tn +1,2-tn +1,1, no motion of the target along the radius direction;
the radius direction is the connecting direction of the target and the target tracking device;
d, judging the size between the step d and the step d, if:
The target moves along the rotation direction of the first image sensor 4 in the vertical radius direction;
The target moves along the vertical radius direction against the rotation direction of the first image sensor 4;
no movement is generated in the direction of vertical radius of the target;
The radius direction is a connecting direction of the target and the target tracking device.
it should be noted that the binocular vision bracket of the present embodiment is implemented on the basis of the first embodiment or the second embodiment.
Detailed description of the invention
The present embodiment is an example of an image-based target swivel determination method.
The target swivel judging method based on the image comprises the following steps:
Step a, when a target enters the first image sensor 4 for the nth time, the second image sensor 7 images to obtain an image f1, and when the target enters the first image sensor 4 for the (n + 1) th time, the second image sensor 7 images to obtain an image f 2;
b, performing 2 × 2 expansion on the image f1 to obtain an image f 3;
step c, calculating according to the following formula
Where M N is the resolution of f1 or image f 2; i and j represent the ith row and jth column of the image, respectively; x is 1,2, …, M, y is 1,2, …, N;
step d, judging whether max [ r (x, y) ] is smaller than a threshold value, if:
It is, the target does not have a swivel;
otherwise, the target has a swivel.
It should be noted that the binocular vision bracket of the present embodiment is implemented on the basis of the first embodiment, the second embodiment or the third embodiment.
Claims (4)
1. the non-three-dimensional imaging binocular vision infrared image target tracking device is characterized by comprising a collimator lens (1) and a semi-reflecting and semi-transmitting lens (2) which are sequentially arranged along a light propagation direction, and further comprising a first imaging objective lens (3) and a first image sensor (4) which are arranged on a transmission light path of the semi-reflecting and semi-transmitting lens (2), wherein an image plane of the first image sensor (4) is overlapped with a focal plane of the first imaging objective lens (3); the imaging device also comprises a reflecting prism (5), a second imaging objective lens (6) and a second image sensor (7) which are arranged on a reflection light path of the half-reflecting and half-transmitting mirror (2), wherein an image plane of the second image sensor (7) is superposed with a focal plane of the second imaging objective lens (6); the first image sensor (4) can rotate at a constant speed by taking the vertical direction as an axis.
2. the binocular vision support is characterized by comprising an upper plate (11) and a lower plate (12) which are sequentially and horizontally arranged from top to bottom, wherein the two sides of the upper plate (11) and the lower plate (12) are connected through a connecting rod (13), a second image sensor (7) is fixedly installed at the top of the upper plate (11), a through hole is formed in the lower plate (12), a bearing (14) which is installed in an interference fit manner with the through hole is arranged in the through hole, a rotating shaft (15) is installed in the bearing (14) in an interference fit manner, the lower part of the rotating shaft (15) is connected with a motor (16) through a coupler, and a first image sensor (4) is installed above the rotating shaft (15) through threads; the motor (16) rotates at a constant speed to drive the rotating shaft (15) to rotate at a constant speed, and further drive the first image sensor (4) to rotate at a constant speed by taking the vertical direction as an axis.
3. a non-three-dimensional imaging binocular vision infrared image target tracking method requires a first image sensor (4) to image under the condition of uniform rotation by taking a vertical direction as an axis, and the rotation angular speed is omega; the second image sensor (7) images under a fixed condition; the method is characterized by comprising the following steps:
step a, judging whether a target has a rotating body by using a second image sensor (7), if so:
if yes, returning to the step a;
If not, entering the step b;
b, recording the time tn,1 when the target enters the field of view of the first image sensor (4) for the nth time and the time tn,2 when the target leaves the field of view of the first image sensor (4) for the nth time; recording the time tn +1,1 when the target enters the field of view of the first image sensor (4) at the n +1 th time and the time tn +1,2 when the target leaves the field of view of the first image sensor (4) at the n +1 th time;
step c, judging the size between tn,2-tn,1 and tn +1,2-tn +1, if:
tn,2-tn,1 > tn +1,2-tn +1,1, the target is far away from the target tracking device along the radius direction;
tn,2-tn,1 > tn +1,2-tn +1,1, the target being close to the target tracking means in the radial direction;
tn,2-tn,1 > tn +1,2-tn +1,1, no motion of the target along the radius direction;
The radius direction is the connecting direction of the target and the target tracking device;
d, judging the size between the step d and the step d, if:
the target moves along the rotating direction of the first image sensor (4) along the vertical radius direction;
the vertical radius direction of the target moves against the rotating direction of the first image sensor (4);
no movement is generated in the direction of vertical radius of the target;
the radius direction is a connecting direction of the target and the target tracking device.
4. An image-based target turning judgment method is characterized by comprising the following steps:
Step a, when a target enters a first image sensor (4) for the nth time, a second image sensor (7) images to obtain an image f1, and when the target enters the first image sensor (4) for the (n + 1) th time, the second image sensor (7) images to obtain an image f 2;
b, performing 2 × 2 expansion on the image f1 to obtain an image f 3;
step c, calculating according to the following formula
where M N is the resolution of f1 or image f 2; i and j represent the ith row and jth column of the image, respectively; x is 1,2, …, M, y is 1,2, …, N;
Step d, judging whether max [ r (x, y) ] is smaller than a threshold value, if:
It is, the target does not have a swivel;
otherwise, the target has a swivel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910845157.7A CN110543015B (en) | 2019-09-08 | 2019-09-08 | Non-three-dimensional imaging binocular vision infrared image target tracking device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910845157.7A CN110543015B (en) | 2019-09-08 | 2019-09-08 | Non-three-dimensional imaging binocular vision infrared image target tracking device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110543015A true CN110543015A (en) | 2019-12-06 |
| CN110543015B CN110543015B (en) | 2022-02-18 |
Family
ID=68712841
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910845157.7A Expired - Fee Related CN110543015B (en) | 2019-09-08 | 2019-09-08 | Non-three-dimensional imaging binocular vision infrared image target tracking device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110543015B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101334276A (en) * | 2007-06-27 | 2008-12-31 | 中国科学院自动化研究所 | Visual sense measurement method and device |
| CN105752580A (en) * | 2016-01-31 | 2016-07-13 | 西北农林科技大学 | Cereal simple-grain dynamical conveying and binocular image collection device |
| CN106227241A (en) * | 2016-08-01 | 2016-12-14 | 西北工业大学 | A kind of motion platform double-view field double loop scan tracing system and method |
| CN106343950A (en) * | 2016-09-28 | 2017-01-25 | 天津工业大学 | Fundus camera binocular stereo-imaging system based on eye model |
| CN108351654A (en) * | 2016-02-26 | 2018-07-31 | 深圳市大疆创新科技有限公司 | System and method for visual target tracking |
-
2019
- 2019-09-08 CN CN201910845157.7A patent/CN110543015B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101334276A (en) * | 2007-06-27 | 2008-12-31 | 中国科学院自动化研究所 | Visual sense measurement method and device |
| CN105752580A (en) * | 2016-01-31 | 2016-07-13 | 西北农林科技大学 | Cereal simple-grain dynamical conveying and binocular image collection device |
| CN108351654A (en) * | 2016-02-26 | 2018-07-31 | 深圳市大疆创新科技有限公司 | System and method for visual target tracking |
| CN106227241A (en) * | 2016-08-01 | 2016-12-14 | 西北工业大学 | A kind of motion platform double-view field double loop scan tracing system and method |
| CN106343950A (en) * | 2016-09-28 | 2017-01-25 | 天津工业大学 | Fundus camera binocular stereo-imaging system based on eye model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110543015B (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5010634B2 (en) | Mobile image tracking device | |
| CA2836697C (en) | System, device and method for tracking position and orientation of vehicle, loading device and cargo in loading device operations | |
| JP4982407B2 (en) | Mobile object image tracking apparatus and method | |
| CN110246175A (en) | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera | |
| US11240436B2 (en) | Machine vision system and method with steerable mirror | |
| FR2912274A1 (en) | Three dimensional/two dimensional virtual or real video camera capturing controlling method, involves generating ocular control signal from detected slow movement, and producing control from ocular control signal | |
| WO2010021114A1 (en) | Long-distance target detection camera system | |
| CN104616322A (en) | Onboard infrared target image identifying and tracking method and device | |
| CN108303078A (en) | A kind of early warning of omnidirection shipping anti-collision and navigation system based on stereoscopic vision | |
| CN105953771A (en) | Active theodolite system and measuring method | |
| WO2018106262A1 (en) | Feature recognition assisted super-resolution method | |
| CN110543015B (en) | Non-three-dimensional imaging binocular vision infrared image target tracking device and method | |
| CN113885312A (en) | Photoelectric tracking system and method | |
| JP2009260564A (en) | Mobile object image tracking apparatus | |
| CN116158780A (en) | Method for carrying out multi-mode ultrasonic imaging on large-size target | |
| JP2686685B2 (en) | An active camera search system for object manipulation by robots | |
| CN113184767B (en) | Aerial work platform navigation method, device and equipment and aerial work platform | |
| CN109993768B (en) | Aerial target spectrum mapping method for improving real-time performance and accuracy of servo tracking | |
| Capurro et al. | A binocular active vision system using space variant sensors: Exploiting autonomous behaviors for space application | |
| WO2021028910A1 (en) | A gimbal apparatus system and method for automated vehicles | |
| KR102544345B1 (en) | Tracking apparatus and method using the same | |
| JPH03174668A (en) | Omnidirectional observing device | |
| Jiang et al. | Multi-Thread AI Cameras Using High-Speed Active Vision System | |
| JP2000266985A (en) | Automatic focal adjustment device for monitor camera | |
| Chang et al. | A Dynamic Target Recognition Method Based on Correlation Tracking Algorithm |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220218 |