CN109035307B - Set area target tracking method and system based on natural light binocular vision - Google Patents
Set area target tracking method and system based on natural light binocular vision Download PDFInfo
- Publication number
- CN109035307B CN109035307B CN201810775921.3A CN201810775921A CN109035307B CN 109035307 B CN109035307 B CN 109035307B CN 201810775921 A CN201810775921 A CN 201810775921A CN 109035307 B CN109035307 B CN 109035307B
- Authority
- CN
- China
- Prior art keywords
- target
- determining
- interest
- image
- information
- 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.)
- Active
Links
Images
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
- G06T7/292—Multi-camera tracking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/285—Analysis of motion using a sequence of stereo image pairs
-
- 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
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
Abstract
The invention discloses a method and a system for tracking a target in a set area based on natural light binocular vision, wherein the method comprises the following steps: acquiring image sequences acquired by two ordinary cameras; determining a constraint condition of an interest target and a depth of field interval set by a user; processing images in the image sequence, and performing an interest target extraction operation on the processed images by adopting a binocular vision tracking algorithm and the constraint condition of the interest target to determine the coordinates of the interest target; determining an image processing interval based on a mapping relation between a world coordinate system of the double cameras and two-dimensional coordinates of the interest target and a depth of field interval set by a user; and processing image sequences acquired by the two ordinary cameras in the image processing interval, determining the three-dimensional coordinates of the interest target in the image processing interval, and realizing target tracking. By applying the method or the system provided by the invention, the image processing interval can be controlled according to actual requirements, and the requirements in privacy protection application scenes are met.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a method and a system for tracking a target in a set area based on natural light binocular vision.
Background
Human beings perceive external information, more than 80% being obtained visually. Vision not only refers to the feeling of light signals, but also includes the processes of acquisition, processing, transmission, storage and understanding of visual information. After the signal processing theory and computer appeared, people tried to acquire an environment image by using a camera and convert the environment image into a digital signal, and the whole process of processing a visual signal is realized by using the computer, so that a new subject-computer vision is formed. Computer vision is to obtain images or image sequences through an image sensor, and then analyze and understand the images through a computer by means of combining technologies such as image processing, mode recognition, artificial intelligence and the like, so as to describe and explain the three-dimensional world. The ultimate goal of computer vision is to replace the human brain in part to achieve real world understanding and comprehension.
In the study of the biological visual system, it was found that almost all living beings having vision have two eyes, and the depth or near-far sensation is given by observing an object with both eyes at the same time. Therefore, in a computer Vision system, one or more cameras are also commonly used to observe the same scene from two or more viewpoints, obtain a set of images at different viewing angles, and then deduce the spatial geometry and position of a target object in the scene through the parallax of the same scene point in different images, which is called stereoscopic Vision (Stereo Vision), and is an important branch of computer Vision and the core content of computer Vision.
The vision system with two cameras is a binocular stereo vision system. The binocular stereoscopic vision process is very similar to the stereoscopic perception process of human vision, and the method for processing the scenery by directly simulating the two eyes of human beings is simple, convenient and reliable. Because of having great application prospect, binocular stereoscopic vision has been paid attention to by scholars in various fields all the time. Since the 21 st century, with the continuous development of research level, stereoscopic vision technology is more and more widely applied to various aspects of social life, such as detection and measurement of industrial products, three-dimensional analysis of medical images, interpretation of aerial photographs and satellite photographs, three-dimensional mapping, and visual navigation of mobile robots.
The moving target detection and tracking technology is another important content of computer vision, and the essence of the technology is that a moving target in a scene is detected and subjected to feature extraction and tracking by analyzing a video sequence shot by a camera, and further the motion parameter of the target is estimated. The moving target detection and tracking integrates advanced technologies in many fields, including image processing, mode recognition, artificial intelligence, automatic control and the like, and is widely applied to many aspects of robot visual navigation, public scene monitoring, military visual guidance, intelligent transportation and the like. The binocular stereo vision simulates a human binocular mechanism, parallax information between stereo image pairs can be extracted, three-dimensional information of an actual scene is recovered to a certain extent, and the binocular stereo vision has irreplaceable advantages when the actual three-dimensional position and depth (namely the distance between a target and an observer) of the target need to be measured. Therefore, the method is of great significance in the fields of accurate tracking, three-dimensional information recovery, target measurement and the like.
At present, images acquired by an infrared camera are mainly used as input of a binocular vision system, and because infrared imaging only has gray information, the defects of incapability of recovering original phases, high equipment cost, high background noise, low resolution and the like are overcome. In addition, current binocular vision systems cannot meet the requirements of application scenarios where privacy protection is emphasized.
Disclosure of Invention
The invention aims to provide a method and a system for tracking a target in a set area based on natural light binocular vision, which can control an image processing interval, meet the requirements in an application scene emphasizing privacy protection, and have the advantages of high target tracking efficiency, low cost, high resolution and the like.
In order to achieve the purpose, the invention provides the following scheme:
a target tracking method for a set area based on natural light binocular vision comprises the following steps:
acquiring image sequences acquired by two ordinary cameras;
determining the constraint condition of the interest target;
processing images in the sequence of images;
performing an interest target extraction operation on the processed image by adopting a binocular vision tracking algorithm and the constraint condition of the interest target, and determining the coordinate of the interest target; the coordinates comprise two-dimensional coordinates and three-dimensional coordinates;
acquiring a depth of field interval set by a user;
determining an image processing interval according to the mapping relation between the world coordinate system of the double cameras and the two-dimensional coordinates of the interest target and the depth of field interval set by the user;
and processing the image sequences acquired by the two ordinary cameras in the image processing interval, and determining the three-dimensional coordinates of the interest target in the image processing interval.
Optionally, processing the images in the image sequence specifically includes:
and carrying out graying, binarization and smooth filtering processing on the images in the image sequence.
Optionally, the performing, by using a binocular vision tracking algorithm and the constraint condition of the interest target, an interest target extraction operation on the processed image specifically includes:
carrying out closed region extraction operation on the processed image, and determining each target object in the processed image;
acquiring characteristic information of each target object; the characteristic information comprises size information, roundness-like information and brightness information;
and comparing the characteristic information of each target object with the constraint conditions of the interest targets to determine the interest targets.
Optionally, the determining the coordinates of the interest target specifically includes:
calibrating a camera and determining camera parameters;
establishing a world coordinate system according to the camera parameters;
determining the coordinates of the interest target according to the world coordinate system and the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
Optionally, the obtaining of the feature information of each target object specifically includes:
extracting the contour line of each target object, and determining roundness-like information and size information of each target object; the dimension information comprises height, width and area;
and determining brightness information of each target object.
Optionally, the processing the image sequence acquired by the two general cameras in the image processing interval to determine the three-dimensional coordinate of the interest target in the image processing interval specifically includes:
and dividing the images in the image sequence according to the image processing interval, determining an image sequence to be processed, returning to the step of processing the images in the image sequence, and replacing the image sequence with the image sequence to be processed until the three-dimensional coordinates of the interest target in the image processing interval are determined.
The invention also provides a set area target tracking system based on natural light binocular vision, which comprises:
the image sequence acquisition module is used for acquiring image sequences acquired by the two ordinary cameras;
the constraint condition determining module is used for determining the constraint condition of the interest target;
the processing module is used for processing the images in the image sequence;
the coordinate determination module is used for extracting the interest target from the processed image by adopting a binocular vision tracking algorithm and the constraint condition of the interest target and determining the coordinate of the interest target; the coordinates comprise two-dimensional coordinates and three-dimensional coordinates;
the field depth interval acquisition module is used for acquiring a field depth interval set by a user;
the image processing interval determining module is used for determining an image processing interval according to the mapping relation between the world coordinate system of the double cameras and the two-dimensional coordinates of the interest target and the depth of field interval set by the user;
and the interest target three-dimensional coordinate determination module in the set range is used for processing image sequences acquired by the two ordinary cameras in the image processing interval and determining the three-dimensional coordinates of the interest target in the image processing interval.
Optionally, the processing module specifically includes:
and the processing unit is used for carrying out graying, binarization and smooth filtering processing on the images in the image sequence.
Optionally, the coordinate determination module specifically includes:
the target object determining unit is used for carrying out closed region extraction operation on the processed image and determining each target object in the processed image;
a characteristic information acquisition unit configured to acquire characteristic information of each of the target objects; the characteristic information comprises size information, roundness-like information and brightness information;
the interest target determining unit is used for comparing the characteristic information of each target object with the constraint condition of the interest target to determine the interest target;
the camera parameter determining unit is used for calibrating the camera and determining camera parameters;
the world coordinate system establishing unit is used for establishing a world coordinate system according to the camera parameters;
the coordinate determination unit is used for determining the coordinate of the interest target according to the world coordinate system and the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
Optionally, the feature information obtaining unit specifically includes:
the roundness-like information and size information determining subunit is used for extracting the contour line of each target object and determining the roundness-like information and size information of each target object; the dimension information comprises height, width and area;
and the brightness information determining unit is used for determining the brightness information of each target object.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention provides a method and a system for tracking a target in a set area based on natural light binocular vision, wherein the method comprises the following steps: acquiring image sequences acquired by two ordinary cameras; determining a constraint condition of an interest target and a depth of field interval set by a user; processing images in the image sequence, and performing an interest target extraction operation on the processed images by adopting a binocular vision tracking algorithm and constraint conditions of an interest target to determine coordinates of the interest target; determining an image processing interval according to the mapping relation between the world coordinate system of the double cameras and the two-dimensional coordinates of the interest target and the depth of field interval set by the user; and processing image sequences acquired by the two ordinary cameras in the image processing interval, and determining the three-dimensional coordinates of the interest target in the image processing interval. By applying the method provided by the invention, the image processing interval can be controlled according to actual requirements, and the requirement in an application scene emphasizing privacy protection is met. In addition, only image pixels in the image processing section are processed, so that the data processing load is reduced, and the real-time performance in the tracking process is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for tracking a target in a set area based on natural-light binocular vision according to an embodiment of the present invention;
FIG. 2 is a comparison chart before and after an anti-interference algorithm is employed in an embodiment of the present invention;
FIG. 3 is a block diagram of a target tracking system for a set area based on natural-light binocular vision according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a target tracking system for a set area based on natural-light binocular vision according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a method and a system for tracking a target in a set area based on natural light binocular vision, which can control depth of field parameters, meet the requirements in an application scene emphasizing privacy protection, and have the advantages of high target tracking efficiency, low cost, high resolution and the like.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a schematic flowchart of a method for tracking a target in a set area based on natural-light binocular vision according to an embodiment of the present invention, and as shown in fig. 1, the method for tracking a target in a set area based on natural-light binocular vision according to an embodiment of the present invention specifically includes the following steps.
Step 101: and acquiring image sequences acquired by two common cameras.
Step 102: constraints of the object of interest are determined.
Step 103: processing images in the sequence of images.
Step 104: performing an interest target extraction operation on the processed image by adopting a binocular vision tracking algorithm and the constraint condition of the interest target, and determining the coordinate of the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
Step 105: and acquiring a depth of field interval set by a user.
Step 106: and determining an image processing interval according to the mapping relation between the world coordinate system of the double cameras and the two-dimensional coordinates of the interest target and the depth of field interval set by the user.
Step 107: and processing the image sequences acquired by the two ordinary cameras in the image processing interval, and determining the three-dimensional coordinates of the interest target in the image processing interval.
Step 103 specifically comprises:
and carrying out graying, binarization and smooth filtering processing on the images in the image sequence.
The embodiment of the invention realizes binocular vision target tracking based on natural light, so that all light sources close to the natural light wavelength are likely to cause interference, and an anti-interference algorithm is adopted for processing.
The target object reflects natural light, and the camera can capture an image of the target object. Objects have their typical characteristics such as width, height, brightness, circularity, etc. Generally, a plurality of features are required to accurately describe a target object. The anti-interference algorithm adopted by the embodiment of the invention is just to identify the interested target through a series of characteristics and filter out non-target objects. In the concrete implementation, the control of the size information such as width, height, area and the like is simple. The control of the brightness information is mainly used for filtering the light source interference, processing the image to obtain a gray level image in the implementation process, reducing the data processing burden, and setting a 'filtering label' for the target higher or lower than a certain gray level (threshold value). The circularity similarity refers to the degree that a target two-dimensional projection tends to be circular, the circularity similarity of the target is calculated through an algorithm in the implementation process, meanwhile, a lower limit and an upper limit are set, a circularity similarity interval for filtering interference is defined, objects or light sources outside the interval are set with 'filtering labels', and finally whether filtering is executed or not is determined according to label weights, which is a main design idea of an anti-interference algorithm. The roundness is a relatively basic and mature calculation process, a target contour is firstly calculated by using a function (such as findcontours), then the radii of a minimum circumcircle and a minimum inscribed circle of the contour are respectively calculated, and the closer the two radii are, the closer the target is to the circle is.
The effectiveness of the anti-interference algorithm is demonstrated by experiments below
The experimental conditions are as follows: under the natural illumination condition, two ordinary cameras are selected as image acquisition equipment, and the luminous balls are used as tracking targets. The stability of an anti-interference algorithm and the stability and the real-time performance of a binocular vision tracking algorithm are mainly detected.
And (3) contrast setting: and comparing the detection and tracking conditions under the condition of existence of the anti-interference algorithm.
The experimental results are as follows: as shown in fig. 2, the left side is the original phase, the middle is the target detection effect of the interference-free algorithm, and the right side is the interference-free algorithm effect. It can be seen that the light sources in the image that interfere with the target have been filtered and the algorithm has tracked the center position of the target (black coordinates in the original phase).
The specific implementation method comprises the following steps: firstly, extracting the outline of each target object in the image (for a computer, the outline represents the object, and each outline corresponds to a number for distinguishing), and further determining the width, the height, the roundness and the like of the target object; the image is converted into a gray scale image, and the brightness of the target object is determined. And comparing the feature information of the target object with the feature of the interest target, if the similarity degree meets the set range, considering the target object as the interest target, and otherwise, deleting the target object.
Therefore, step 104 specifically includes:
and carrying out closed region extraction operation on the processed image, and determining each target object in the processed image.
Acquiring characteristic information of each target object; the characteristic information comprises size information, roundness-like information and brightness information; specifically, the contour line of each target object is extracted, and roundness-like information and size information of each target object are determined; the dimension information comprises height, width and area; and determining brightness information of each target object.
And comparing the characteristic information of each target object with the constraint conditions of the interest targets to determine the interest targets.
And calibrating the camera and determining camera parameters.
And establishing a world coordinate system according to the camera parameters.
Determining the coordinates of the interest target according to the world coordinate system and the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
Step 106 specifically includes:
one important advantage of binocular vision systems is that depth information of objects within the field of view can be obtained, and in order to make full use of this, depth-of-field control algorithms are designed. The image processing interval can be controlled from the perspective by the depth-of-field control algorithm, objects outside the depth-of-field interval cannot be processed, and the method has certain significance for application scenes with space protection requirements (such as privacy protection problems of public monitoring systems).
The three-dimensional coordinates of the interest target are provided with depth information, when a depth-of-field control algorithm is executed, a plurality of interest target points are collected at a certain row distance or column distance for each frame of image in the common visual field of the two cameras, three-dimensional data of the interest target points are obtained through a three-dimensional positioning process, wherein the depth information is contained, the mapping relation based on the three-dimensional coordinates and the two-dimensional coordinates of the cameras is drawn by the aid of the plurality of interest target points, then a depth-of-field control interval preset by a user is read, and the mapping relation is fed back to the two-dimensional image processing unit according to the drawing, so that image processing control of the set.
The specific mapping relationship depends on the application environment and the device used, because the result of camera calibration is used. The so-called "three-dimensional coordinates" is a coordinate system (also called a world coordinate system) based on the two cameras, which is established after the cameras are calibrated, then three-dimensional coordinates in the world coordinate system are calculated according to a set of two-dimensional coordinates of the same point acquired by the two cameras, and the three-dimensional coordinates correspond to one another, and further a mapping relation between the two-dimensional coordinates and the three-dimensional coordinates is established.
Step 107 specifically includes:
and dividing the images in the image sequence according to the image processing interval, determining an image sequence to be processed, returning to the step 102, and replacing the image sequence with the image sequence to be processed until the three-dimensional coordinates of the interest target in the image processing interval are determined.
Fig. 3 is a block diagram of a target tracking system in a set area based on natural-light binocular vision according to an embodiment of the present invention, and as shown in fig. 3, the basic devices of the target tracking system in the set area are two general cameras. The method comprises the steps that a camera collects images in a visual field, graying, binarization, smooth filtering and other processing are carried out at the image processing stage, then an anti-interference algorithm extracts object contours, targets are identified and non-target objects are filtered according to preset features such as size information, brightness and roundness, pixel matching is carried out on each frame according to the features, and tracking of the targets is sequentially kept. And in the three-dimensional positioning stage, the three-dimensional coordinates of the current frame target are calculated according to the established three-dimensional coordinate system and the two-dimensional image obtained in the tracking process, so that the positioning is completed. And then realizing the depth of field control according to the depth of field parameter set by the user, returning to the target image processing unit, further realizing the target image processing within the set range, and finishing the target positioning and tracking within the set range.
Fig. 4 is a schematic structural diagram of a target tracking system for a set area based on natural-light binocular vision according to an embodiment of the present invention, and as shown in fig. 4, the target tracking system for a set area based on natural-light binocular vision according to an embodiment of the present invention includes:
an image sequence acquiring module 100, configured to acquire image sequences acquired by two general cameras.
A constraint condition determining module 200, configured to determine a constraint condition of the object of interest.
A processing module 300, configured to process the images in the image sequence.
A coordinate determination module 400, configured to perform an interest target extraction operation on the processed image by using a binocular vision tracking algorithm and a constraint condition of the interest target, and determine a coordinate of the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
The depth-of-field interval acquiring module 500 is configured to acquire a depth-of-field interval set by a user.
And the image processing interval determining module 600 is configured to determine an image processing interval according to a mapping relationship between a world coordinate system of the two cameras and the two-dimensional coordinate of the interest target and a depth of field interval set by a user.
And the interest target three-dimensional coordinate determination module 700 in the set range is configured to process image sequences acquired by the two general cameras in the image processing section, and determine the three-dimensional coordinate of the interest target in the image processing section.
The processing module 300 specifically includes:
and the processing unit is used for carrying out graying, binarization and smooth filtering processing on the images in the image sequence.
The coordinate determination module 400 specifically includes:
and the target object determining unit is used for carrying out closed region extraction operation on the processed image and determining each target object in the processed image.
A characteristic information acquisition unit configured to acquire characteristic information of each of the target objects; the characteristic information comprises size information, roundness-like information and brightness information.
And the interest target determining unit is used for comparing the characteristic information of each target object with the constraint condition of the interest target and determining the interest target.
And the camera parameter determining unit is used for calibrating the camera and determining camera parameters.
And the world coordinate system establishing unit is used for establishing a world coordinate system according to the camera parameters.
The coordinate determination unit is used for determining the coordinate of the interest target according to the world coordinate system and the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
The feature information acquiring unit specifically includes:
the roundness-like information and size information determining subunit is used for extracting the contour line of each target object and determining the roundness-like information and size information of each target object; the dimensional information includes height, width, and area.
And the brightness information determining unit is used for determining the brightness information of each target object.
The invention can not play a role in some practical applications with high performance requirements due to the fact that the traditional binocular vision tracking system based on feature training cannot effectively balance real-time performance and accuracy, and needs to manually frame a target in a first frame, and is not intelligent and flexible enough in practical applications, and under the background that public video monitoring and privacy protection are the hot points of current social discussion, the target of interest is tracked in real time in a set range by adopting an anti-interference algorithm and a depth-of-field control algorithm, and public privacy is well protected.
Compared with the prior art, the invention has the advantages that:
1. compared with an infrared camera, the method has the advantages that the common camera is adopted as the acquisition equipment, and the original phase can be obtained for processing and displaying, so that the cost is reduced, and the requirement of displaying the original phase is met.
2. The interference degree of a binocular system based on natural light is reduced to a certain extent by adopting an anti-interference algorithm, and the effectiveness of the anti-interference algorithm is proved by experiments.
3. The image processing range can be controlled according to actual requirements, the complexity of feature extraction and the data processing burden are simplified, the real-time performance of the tracking process is improved, and the method has certain significance for application scenes with space protection requirements (such as privacy protection problems of public monitoring systems).
4. The collected image can be displayed, and the image processed in the image processing section can be displayed.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (9)
1. A target tracking method for a set area based on natural light binocular vision is characterized by comprising the following steps:
acquiring image sequences acquired by two ordinary cameras;
determining the constraint condition of the interest target;
processing images in the sequence of images;
performing an interest target extraction operation on the processed image by adopting a binocular vision tracking algorithm and the constraint condition of the interest target, and determining the coordinate of the interest target; the coordinates comprise two-dimensional coordinates and three-dimensional coordinates;
acquiring a depth of field interval set by a user;
determining an image processing interval according to the mapping relation between the world coordinate system of the double cameras and the two-dimensional coordinates of the interest target and the depth of field interval set by the user;
processing image sequences acquired by two ordinary cameras in the image processing interval, and determining three-dimensional coordinates of the interest target in the image processing interval, specifically comprising: and dividing the images in the image sequence according to the image processing interval, determining an image sequence to be processed, returning to the step of processing the images in the image sequence, and replacing the image sequence with the image sequence to be processed until the three-dimensional coordinates of the interest target in the image processing interval are determined.
2. The method for tracking the target in the set area according to claim 1, wherein processing the images in the image sequence specifically comprises:
and carrying out graying, binarization and smooth filtering processing on the images in the image sequence.
3. The method for tracking the target in the set area according to claim 1, wherein the extracting the target of interest from the processed image by using a binocular vision tracking algorithm and a constraint condition of the target of interest specifically comprises:
carrying out closed region extraction operation on the processed image, and determining each target object in the processed image;
acquiring characteristic information of each target object; the characteristic information comprises size information, roundness-like information and brightness information;
and comparing the characteristic information of each target object with the constraint conditions of the interest targets to determine the interest targets.
4. The method for tracking a target within a defined area as claimed in claim 3, wherein said determining the coordinates of the target of interest comprises
Calibrating a camera and determining camera parameters;
establishing a world coordinate system according to the camera parameters;
determining the coordinates of the interest target according to the world coordinate system and the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
5. The method for tracking targets in a set area according to claim 3, wherein the acquiring the feature information of each target object specifically includes:
extracting the contour line of each target object, and determining roundness-like information and size information of each target object; the dimension information comprises height, width and area;
and determining brightness information of each target object.
6. A set-zone target tracking system based on natural-light binocular vision, the set-zone target tracking system comprising:
the image sequence acquisition module is used for acquiring image sequences acquired by the two ordinary cameras;
the constraint condition determining module is used for determining the constraint condition of the interest target;
the processing module is used for processing the images in the image sequence;
the coordinate determination module is used for extracting the interest target from the processed image by adopting a binocular vision tracking algorithm and the constraint condition of the interest target and determining the coordinate of the interest target; the coordinates comprise two-dimensional coordinates and three-dimensional coordinates;
the field depth interval acquisition module is used for acquiring a field depth interval set by a user;
the image processing interval determining module is used for determining an image processing interval according to the mapping relation between the world coordinate system of the double cameras and the two-dimensional coordinates of the interest target and the depth of field interval set by the user;
an interest target three-dimensional coordinate determination module within a set range, configured to process image sequences acquired by two general cameras within the image processing interval, and determine a three-dimensional coordinate of the interest target within the image processing interval, specifically including: and dividing the images in the image sequence according to the image processing interval, determining an image sequence to be processed, returning to the processing module, and replacing the image sequence with the image sequence to be processed until the three-dimensional coordinates of the interest target in the image processing interval are determined.
7. The system for tracking targets in a set area according to claim 6, wherein the processing module specifically comprises:
and the processing unit is used for carrying out graying, binarization and smooth filtering processing on the images in the image sequence.
8. The system for tracking targets within a defined area as claimed in claim 6, wherein the coordinate determination module comprises:
the target object determining unit is used for carrying out closed region extraction operation on the processed image and determining each target object in the processed image;
a characteristic information acquisition unit configured to acquire characteristic information of each of the target objects; the characteristic information comprises size information, roundness-like information and brightness information;
the interest target determining unit is used for comparing the characteristic information of each target object with the constraint condition of the interest target to determine the interest target;
the camera parameter determining unit is used for calibrating the camera and determining camera parameters;
the world coordinate system establishing unit is used for establishing a world coordinate system according to the camera parameters;
the coordinate determination unit is used for determining the coordinate of the interest target according to the world coordinate system and the interest target; the coordinates include two-dimensional coordinates and three-dimensional coordinates.
9. The system for tracking a target in a set area according to claim 8, wherein the characteristic information acquiring unit specifically includes:
the roundness-like information and size information determining subunit is used for extracting the contour line of each target object and determining the roundness-like information and size information of each target object; the dimension information comprises height, width and area;
and the brightness information determining unit is used for determining the brightness information of each target object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810775921.3A CN109035307B (en) | 2018-07-16 | 2018-07-16 | Set area target tracking method and system based on natural light binocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810775921.3A CN109035307B (en) | 2018-07-16 | 2018-07-16 | Set area target tracking method and system based on natural light binocular vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109035307A CN109035307A (en) | 2018-12-18 |
| CN109035307B true CN109035307B (en) | 2020-09-25 |
Family
ID=64643062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810775921.3A Active CN109035307B (en) | 2018-07-16 | 2018-07-16 | Set area target tracking method and system based on natural light binocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109035307B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110197508B (en) * | 2019-07-10 | 2024-02-20 | 深圳西顺万合科技有限公司 | 2D and 3D co-fusion vision guiding movement method and device |
| CN110992393B (en) * | 2019-11-24 | 2023-06-30 | 思看科技(杭州)股份有限公司 | Target motion tracking method based on vision |
| CN111953933B (en) * | 2020-07-03 | 2022-07-05 | 北京中安安博文化科技有限公司 | Method, device, medium and electronic equipment for determining fire area |
| CN113516709B (en) * | 2021-07-09 | 2023-12-29 | 连云港远洋流体装卸设备有限公司 | Flange positioning method based on binocular vision |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108234920A (en) * | 2012-12-20 | 2018-06-29 | 微软技术许可有限责任公司 | Video camera with privacy mode |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101308018B (en) * | 2008-05-30 | 2010-09-15 | 汤一平 | Stereo vision measuring apparatus based on binocular omnidirectional visual sense sensor |
| CN101883291B (en) * | 2010-06-29 | 2012-12-19 | 上海大学 | Method for drawing viewpoints by reinforcing interested region |
| CN102135236B (en) * | 2011-01-05 | 2013-05-01 | 北京航空航天大学 | Automatic non-destructive testing method for internal wall of binocular vision pipeline |
| CN102699733B (en) * | 2012-06-12 | 2014-06-25 | 大连理工大学 | Method and device for measuring movement locus of automatic tool changing mechanical arm |
| US10049462B2 (en) * | 2016-03-23 | 2018-08-14 | Akcelita, LLC | System and method for tracking and annotating multiple objects in a 3D model |
| US20180041747A1 (en) * | 2016-08-03 | 2018-02-08 | Samsung Electronics Co., Ltd. | Apparatus and method for processing image pair obtained from stereo camera |
| CN107909604A (en) * | 2017-11-07 | 2018-04-13 | 武汉科技大学 | Dynamic object movement locus recognition methods based on binocular vision |
-
2018
- 2018-07-16 CN CN201810775921.3A patent/CN109035307B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108234920A (en) * | 2012-12-20 | 2018-06-29 | 微软技术许可有限责任公司 | Video camera with privacy mode |
Non-Patent Citations (1)
| Title |
|---|
| Focal Track: Depth and Accommodation with Oscillating Lens Deformation;Qi Guo等;《2017 IEEE International Conference on Computer Vision》;20171225;第966-973页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109035307A (en) | 2018-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110221690B (en) | Gesture interaction method and device based on AR scene, storage medium and communication terminal | |
| CN109035307B (en) | Set area target tracking method and system based on natural light binocular vision | |
| CN110458025B (en) | Target identification and positioning method based on binocular camera | |
| Fang et al. | Light filed image quality assessment by local and global features of epipolar plane image | |
| US20160093101A1 (en) | Method And System For Generating A Three-Dimensional Model | |
| CN112528974B (en) | Distance measuring method and device, electronic equipment and readable storage medium | |
| CN113378649A (en) | Identity, position and action recognition method, system, electronic equipment and storage medium | |
| CN113256731A (en) | Target detection method and device based on monocular vision | |
| CN111127556B (en) | Target object identification and pose estimation method and device based on 3D vision | |
| CN108230351A (en) | Sales counter evaluation method and system based on binocular stereo vision pedestrian detection | |
| CN110909571B (en) | High-precision face recognition space positioning method | |
| CN115035546A (en) | Three-dimensional human body posture detection method and device and electronic equipment | |
| CN114511608A (en) | Method, device, terminal, imaging system and medium for acquiring depth image | |
| CN111862511B (en) | Target intrusion detection device and method based on binocular stereo vision | |
| CN112073640B (en) | Panoramic information acquisition pose acquisition method, device and system | |
| CN107145820B (en) | Binocular positioning method based on HOG characteristics and FAST algorithm | |
| CN103006332A (en) | Scalpel tracking method and device and digital stereoscopic microscope system | |
| CN106651921B (en) | Motion detection method and method for avoiding and tracking moving target | |
| KR101520293B1 (en) | Scheduling method for detention of object and of obtaining forensic image by visual attention, and system thereof | |
| CN116862832A (en) | Three-dimensional live-action model-based operator positioning method | |
| Hadi et al. | Fusion of thermal and depth images for occlusion handling for human detection from mobile robot | |
| CN112099330B (en) | Holographic human body reconstruction method based on external camera and wearable display control equipment | |
| KR20160039447A (en) | Spatial analysis system using stereo camera. | |
| KR101668649B1 (en) | Surrounding environment modeling method and apparatus performing the same | |
| CN107802468B (en) | Blind guiding method and blind guiding system |
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 |