CN115784026A - Wavefront sensing and imaging combined type three-dimensional tracking system and method for positioning lifting hook - Google Patents
Wavefront sensing and imaging combined type three-dimensional tracking system and method for positioning lifting hook Download PDFInfo
- Publication number
- CN115784026A CN115784026A CN202310058033.0A CN202310058033A CN115784026A CN 115784026 A CN115784026 A CN 115784026A CN 202310058033 A CN202310058033 A CN 202310058033A CN 115784026 A CN115784026 A CN 115784026A
- Authority
- CN
- China
- Prior art keywords
- imaging
- light source
- dimensional
- infrared
- vertical surface
- 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 80
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 30
- 230000011218 segmentation Effects 0.000 claims abstract description 3
- 230000003287 optical effect Effects 0.000 claims description 18
- 238000013527 convolutional neural network Methods 0.000 claims description 17
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 9
- 229910001316 Ag alloy Inorganic materials 0.000 claims description 4
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 claims description 4
- 150000001875 compounds Chemical class 0.000 claims 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 39
- 230000008901 benefit Effects 0.000 abstract description 4
- 238000009434 installation Methods 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 description 20
- 238000010276 construction Methods 0.000 description 16
- 230000000007 visual effect Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000002834 transmittance Methods 0.000 description 3
- FAPWRFPIFSIZLT-UHFFFAOYSA-M Sodium chloride Chemical compound [Na+].[Cl-] FAPWRFPIFSIZLT-UHFFFAOYSA-M 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013329 compounding Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003302 ferromagnetic material Substances 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000003116 impacting effect Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000005389 magnetism Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 239000011780 sodium chloride Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
Images
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to the technical field of hook positioning, in particular to a wavefront sensing and imaging combined type three-dimensional tracking system and method for hook positioning. The system comprises: the beacon light source is arranged at the lifting hook and used for generating an infrared light signal; the infrared wavefront sensing unit is used for acquiring light spot information of the infrared light signal, and the light spot information is array light spots obtained after the infrared light signal is subjected to segmentation processing; the visible light imaging unit is used for acquiring a plane image at the working area of the lifting hook, and the imaging surface of the visible light imaging unit is a vertical surface; and the data processing unit acquires and acquires alarm information according to the light spot information and the plane image. The method is realized based on the system. The invention has the advantages of small volume, large detection range, high detection precision, no need of calibration, convenient installation and use and the like.
Description
Technical Field
The invention relates to the technical field of hook positioning, in particular to a wavefront sensing and imaging combined type three-dimensional tracking system and method for hook positioning.
Background
In the hoisting operation of electric power construction, such as the reconstruction and extension of a transformer substation, the hoisting operation usually involves the operation near a live body, and the hoisting operation is usually located in a dangerous area such as high-voltage live. Currently, when such work is performed, special personnel are usually arranged on a construction site to perform the border crossing warning in a visual mode. Considering that the height of a hoisted object is generally higher and the noise of a construction site is more, the method of simply depending on the manual observation and the warning has the defects that the visual error of the visual observation is uncontrollable, the warning instruction is difficult to be transmitted in time, and the like.
In fact, there are also a large number of localization techniques in the prior art, which can be generally divided into sensing-based techniques and image-based techniques. However, the prior art is difficult to be directly applied to hoisting operation. The reason is as follows:
1. positioning technologies based on induction technology generally include infrared correlation positioning, microwave correlation positioning, UWB positioning, radar ranging, and the like;
the infrared correlation positioning technology, the microwave correlation positioning technology and the like have high installation requirements on detection equipment and limited detection range, and when the method is applied to hoisting operation, the distance between a hoisting working area and the ground is usually more than 10 meters, so that the comprehensive coverage of a construction area is difficult to realize;
in the UWB positioning technology and the like, a large number of complex terminal induction devices need to be deployed in a detection area, and positioning of an object to be positioned (a person or an object) is realized by arranging a tag at the position of the object to be positioned; in fact, hoisting work has the characteristics of short construction period, strong randomness and the like, and due to the cost considerations of construction period, cost and the like, the positioning technology lacks good practical feasibility; in particular, the positioning technology has lower precision, and is difficult to realize the precise positioning of a lifting hook, a lifting object at the lifting hook and the like;
although the common radar ranging technology can better perform local tracking, the common radar ranging technology can only obtain one-dimensional distance information, so that the hoisting area is difficult to be spatially positioned; although there is a three-dimensional lidar technology, which can realize three-dimensional positioning by galvanometer scanning or area array laser, because the complexity of the detection system is high, such a positioning technology lacks better practical feasibility due to the cost of construction period, cost, etc.;
2. the positioning technology based on the image technology generally comprises a common visual detection technology, a binocular visual detection technology, a structured light three-dimensional imaging technology and the like;
the information acquired by the common visual inspection technology is two-dimensional plane image information, and the three-dimensional positioning capability of a hoisting working area is poor;
although the binocular vision detection technology has three-dimensional space positioning capability, the positioning accuracy is poor;
although the structured light three-dimensional imaging technology can acquire high-precision three-dimensional images of local areas, complex system calibration is required each time, and once relevant parameters change in the construction process, the calibration needs to be carried out again, so that the structured light three-dimensional imaging technology is difficult to be directly applied to hoisting operation.
Disclosure of Invention
The invention provides a wavefront sensing and imaging combined type three-dimensional tracking system for positioning a lifting hook, which can overcome the defect that the positioning cost is high or the precision is low so that the system is not suitable for electric power construction hoisting operation in the prior art, and has the advantages of small volume, large detection range, high detection precision, no need of calibration, convenience in installation and use and the like.
The invention provides a wavefront sensing and imaging combined type three-dimensional tracking system for positioning a lifting hook, which comprises:
the beacon light source is arranged at the lifting hook and used for generating an infrared light signal;
the infrared wavefront sensing unit is used for acquiring light spot information of the infrared light signal, and the light spot information is array light spots obtained after the infrared light signal is subjected to segmentation processing;
the visible light imaging unit is used for acquiring a plane image at the working area of the lifting hook, and the imaging surface of the visible light imaging unit is a vertical surface; and (c) a second step of,
and the data processing unit is used for acquiring space coordinate information of the beacon light source based on the light spot information, acquiring a two-dimensional coordinate of the hoisting object at the lifting hook on the vertical surface based on the space coordinate information of the beacon light source and the plane image, and acquiring alarm information based on the two-dimensional coordinate of the obstacle at the working area of the lifting hook on the vertical surface and the two-dimensional coordinate of the hoisting object on the vertical surface.
According to the invention, through the beacon light source and the infrared wavefront sensing unit, the three-dimensional positioning of the beacon light source can be better realized in a wavefront detection mode; it can be understood that, in practical use, the beacon light source can be disposed at the hook, so that the three-dimensional positioning of the hook can be realized through the three-dimensional positioning of the beacon light source. Meanwhile, the position relation of the beacon light source and the hoisted object in the plane image can be obtained through the visible light imaging unit, so that the two-dimensional coordinate of the hoisted object in the vertical plane can be better obtained based on the obtained space coordinate information of the beacon light source, and the border-crossing alarm can be better realized based on the two-dimensional coordinate of the obstacle on the vertical plane based on the two-dimensional coordinate.
Preferably, the infrared wavefront sensing unit and the visible light imaging unit share the same wide-angle lens unit, and the wide-angle lens unit is provided with a transmission optical path and a reflection optical path; the transmission light path is used for transmitting the infrared light signal to the infrared wavefront sensing unit, and the reflection light path is used for transmitting the visible light imaging signal to the visible light imaging unit.
By the means, a wide-angle lens unit is introduced, so that a large detection range can be obtained better; meanwhile, the wide-angle lens unit is a shared device, so that the volume of the detection terminal can be preferably reduced.
Preferably, the infrared wavefront sensing unit is provided with an aplanatic collimating lens, an infrared narrow-band filter, a micro-lens array and an infrared CCD image sensor which are sequentially arranged. The acquisition of the spot information can be preferably achieved.
Preferably, the visible light imaging unit includes a visible light CCD image sensor. Therefore, a plane image of the working area of the lifting hook on the vertical plane can be obtained better.
Preferably, the data processing unit constructs a convolutional neural network model, and the convolutional neural network model is used for inputting the light spot information and outputting the space coordinate information of the beacon light source. Therefore, compared with the traditional centroid algorithm, the centroid algorithm and the like, the method has higher response speed.
Preferably, a coordinate conversion model is constructed at the data processing unit, and the coordinate conversion model is used for inputting the two-dimensional coordinates of the beacon light source on the vertical surface and outputting the two-dimensional coordinates of the hoisted object on the vertical surface; in particular to a method for preparing a high-purity sodium chloride solution,
wherein,
and
respectively the coordinates of the beacon light source in the horizontal direction and the vertical direction on the vertical plane,
and
respectively are the coordinates of the hoisted object in the horizontal direction and the vertical direction on the vertical surface,
and
respectively the coordinates of the beacon light source in the horizontal direction and the vertical direction in the plane image,
and
respectively the coordinates of the hoisted object in the horizontal direction and the vertical direction in the plane image "
"is the product operation.
By the method, the two-dimensional coordinate of the hoisted object can be acquired based on the space coordinate information of the position (the lifting hook position) of the beacon light source.
Preferably, the data processing unit is used for generating and outputting an alarm signal when any one of the horizontal direction coordinates and the vertical direction coordinates of the hoisted object on the vertical surface exceeds the coordinates of the obstacle in the corresponding direction on the vertical surface. Therefore, the lifting hook and the lifting object can be better monitored and alarmed.
In addition, the invention also provides a wave-front sensing and imaging combined type three-dimensional tracking method for hook positioning, which comprises the following steps:
arranging a beacon light source at the lifting hook, wherein the beacon light source is used for generating an infrared light signal;
acquiring light spot information of the infrared light signal through an infrared wavefront sensing unit, wherein the light spot information is array light spots of the infrared light signal after being segmented;
acquiring a plane image at the working area of the lifting hook through a visible light imaging unit, wherein the imaging surface of the plane image is a vertical surface;
the method comprises the steps of obtaining space coordinate information of a beacon light source based on light spot information through a data processing unit, obtaining a two-dimensional coordinate of a hoisting object on a vertical surface at a lifting hook based on the space coordinate information of the beacon light source and a plane image, and obtaining alarm information based on the two-dimensional coordinate of an obstacle on the vertical surface at a working area of the lifting hook and the two-dimensional coordinate of the hoisting object on the vertical surface.
According to the invention, through the wavefront sensing and imaging combined type three-dimensional tracking system, the actual three-dimensional coordinate of the beacon light source can be directly recovered from the light spot information, the actual two-dimensional position of the hoisted object can be directly calculated and obtained according to the plane image, and the cross-border alarm detection of the lifting hook and the hoisted object can be better realized through comparing the actual two-dimensional position with the two-dimensional position of the obstacle. The means can better avoid calibration work, has simple and convenient integral structure and convenient arrangement, and can be better suitable for hoisting operation of electric power construction.
Preferably, a convolutional neural network model is constructed at the data processing unit, the convolutional neural network model being used for inputting the light spot information and for outputting the spatial coordinate information of the beacon light source. Therefore, the acquisition of the space coordinate information of the beacon light source can be better realized, and calibration is not needed.
Preferably, a coordinate conversion model is constructed at the data processing unit, and the coordinate conversion model is used for inputting the two-dimensional coordinates of the beacon light source on the vertical surface and outputting the two-dimensional coordinates of the hoisted object on the vertical surface; in particular to a method for preparing a high-performance nano-silver alloy,
wherein,
and
respectively the coordinates of the beacon light source in the horizontal direction and the vertical direction on the vertical plane,
and
respectively are the coordinates of the hoisted object in the horizontal direction and the vertical direction on the vertical surface,
and
respectively the coordinates of the beacon light source in the horizontal direction and the vertical direction in the plane image,
and
respectively the coordinates of the hoisted object in the horizontal direction and the vertical direction in the plane image "
"is the product operation. Therefore, the two-dimensional coordinate information of the hoisted object can be acquired better.
Drawings
Fig. 1 is a schematic view of a three-dimensional tracking system in embodiment 1;
fig. 2 is a light spot information diagram of the beacon light source S1 in embodiment 1 at different angles of view and different axial distances in the optical axis direction;
fig. 3 is a schematic view of an infrared wavefront sensing optical path system in embodiment 1;
FIG. 4 is a schematic view of a visible light imaging optical path system in embodiment 1;
FIG. 5 is a schematic flowchart of the convolutional neural network model in example 1;
FIG. 6 is a schematic flowchart of a test sample of the convolutional neural network model obtained in example 1;
FIG. 7 is a schematic view of a plane image at the working area of the hook in example 1;
FIG. 8 is a schematic view showing the imaging of a plane image in example 1;
FIG. 9 is a schematic side view showing the disposition of the detection terminal in embodiment 1;
fig. 10 is a schematic top view of the arrangement of the detection terminals in embodiment 1.
Detailed Description
For a further understanding of the contents of the present invention, reference will now be made in detail to the following examples. It is to be understood that the examples are illustrative of the invention and not limiting.
Example 1
In consideration of the fact that in the hoisting operation of electric power construction, local areas of a crane, such as a hook and a hoisting object, need to be tracked and positioned to avoid impacting electrified bodies or obstacles, such as telegraph poles and electric wires, the embodiment provides a wavefront sensing and imaging combined type three-dimensional tracking system for positioning the hook and a corresponding method. Compared with the prior art, the scheme of the invention has the advantages of small volume, large detection range, high detection precision, no need of calibration, convenience in installation and use and the like, can accurately track local positions of a lifting hook, a lifting object and the like simultaneously, adopts a convolutional neural network architecture for data processing, and has high response speed.
Referring to fig. 1, the wavefront sensing and imaging combined three-dimensional tracking system for hook positioning in the present embodiment includes:
a beacon light source S1, which is used for being arranged at the hook and generating an infrared light signal;
the infrared wavefront sensing unit S2 is used for acquiring light spot information of the infrared light signal, wherein the light spot information is an array light spot of the infrared light signal after being segmented;
the visible light imaging unit S3 is used for acquiring a plane image at the working area of the lifting hook, and the imaging surface of the visible light imaging unit S3 is a vertical surface; and (c) a second step of,
and the data processing unit S4 is used for acquiring the space coordinate information of the beacon light source S1 based on the light spot information, acquiring the two-dimensional coordinate of the hoisting object at the lifting hook on the vertical surface based on the space coordinate information of the beacon light source S1 and the plane image, and acquiring alarm information based on the two-dimensional coordinate of the obstacle at the working area of the lifting hook on the vertical surface and the two-dimensional coordinate of the hoisting object on the vertical surface.
In this embodiment, through the beacon light source S1 and the infrared wavefront sensing unit S2, the three-dimensional positioning of the beacon light source S1 can be preferably achieved in a wavefront detection manner; it can be understood that, in practical use, the beacon light source S1 can be disposed at the hook, so that the three-dimensional positioning of the hook can be realized by the three-dimensional positioning of the beacon light source S1. Meanwhile, the position relation between the beacon light source S1 and the hoisted object in the plane image can be obtained through the visible light imaging unit S3, so that the two-dimensional coordinate of the hoisted object in the vertical plane can be better obtained based on the obtained space coordinate information of the beacon light source S1, and the border-crossing alarm can be better realized based on the two-dimensional coordinate of the obstacle on the vertical plane based on the two-dimensional coordinate.
Referring to fig. 2, it can be understood that, for an ideal plane wave emitted from the optical system, the focused light spot after passing through the microlens array will be at the origin of each sub-region, and when there is aberration in the optical system to be measured, the wavefront of the sub-aperture with different slopes enters the microlens array, and the focus point will deviate from the origin after passing through the microlens array. When the infrared light signals have different incident angles, the light spot information of the array light spot form after the division processing is different; that is, the spatial three-dimensional coordinate of the beacon light source S1 and the pattern of the spot information show a one-to-one mapping relationship, so that the spatial coordinate information of the beacon light source S1 can be preferably acquired based on the spot information.
Referring to fig. 2, the light spot information of the beacon light source S1 at different angles of view and different axial distances in the optical axis direction is shown.
The space coordinate information of the beacon light source S1 is acquired through the light spot information, so that a calibration link is omitted, the infrared wavefront sensing unit S2 and the visible light imaging unit S3 in the whole detection system are mostly composed of light path systems, and the data processing unit S4 can be realized by the existing upper computer such as a computer, so that the size of the whole detection part is small, the device can be better suitable for hoisting operation of power construction, and the feasibility of better practical application is realized.
The positioning and tracking overall process of the scheme of the invention comprises the steps of firstly obtaining a three-dimensional space coordinate of a position (namely a hook position) where a beacon light source S1 is located, wherein the three-dimensional space coordinate can be an actual coordinate in a world coordinate system; then, based on the position relation between the hoisted object and the beacon light source S1 in the plane image, the actual two-dimensional coordinate (on the vertical surface) of the hoisted object under the world coordinate system can be better obtained; finally, the two-dimensional coordinate on the vertical surface of the lifting hook and the two-dimensional coordinate on the vertical surface of the lifting object are compared with the two-dimensional coordinate of the obstacle on the vertical surface, so that the alarm information can be acquired better, and the positioning of the lifting hook and the lifting object and the detection of whether the lifting hook and the lifting object cross the border or not can be better realized.
It can be understood that, in theory, a plurality of beacon light sources S1 can be arranged at the lifting hook and the lifting object, and the acquisition of the spatial three-dimensional coordinates of the lifting hook and the lifting object can be realized only by the infrared wavefront sensing unit S2, and meanwhile, the acquisition of the alarm information in the three-dimensional space can be realized by comparing the spatial three-dimensional coordinates with the three-dimensional coordinates of the obstacle in the working area; however, in practical operation, the hoisted object needs to be hoisted to a predetermined position in the air, so that when the hoisted object is hoisted in place, the beacon light source S1 at the position of the hoisted object is difficult to detach. Therefore, in the embodiment, the visible light imaging unit S3 is added to obtain the two-dimensional coordinates of the hoisted object on the vertical plane.
Furthermore, although at the working area the obstacles are spatially distributed, i.e. obstacles may be present in the length, height and width directions of the hook; however, spatial positioning can be achieved by providing a plurality of sets of detection terminals (the whole optical path detection formed by the infrared wavefront sensing unit S2, the visible light imaging unit S3, and the like, the same applies below). If only 1 beacon light source S1 is arranged at the lifting hook, and 2 groups of detection terminals consisting of the infrared wavefront sensing unit S2 and the visible light imaging unit S3 are arranged at the ground, the imaging surfaces of the 2 groups of detection terminals can be orthogonal vertical surfaces, so that the tracking and positioning of the position of the hoisted object in space and the out-of-range alarm detection can be better realized.
In this embodiment, the beacon light source S1 can have an infrared light signal generator (such as an infrared LED) and a base with strong magnetism, so that the beacon light source S1 can be conveniently and rapidly disposed at the hook made of ferromagnetic material. In addition, the wavelength of the infrared light signal generator can be selected to be 1550nm, so that visual pollution to a construction area is avoided. Meanwhile, the infrared LED is adopted as a beacon light source, and the advantage of low power consumption can be achieved.
Referring to fig. 2, the infrared wavefront sensing unit S2 and the visible light imaging unit S3 in this embodiment share the same wide-angle lens unit S2-3, and the wide-angle lens unit S2-3 has a transmission optical path and a reflection optical path; the transmission optical path is used for transmitting the infrared light signal to the infrared wavefront sensing unit S2, and the reflection optical path is used for transmitting the visible light imaging signal to the visible light imaging unit S3.
By the means, the wide-angle lens unit S2-3 can be introduced to better obtain a larger detection range, and as a specific scheme, the half field angle of the embodiment is 50 degrees; meanwhile, the wide-angle lens unit S2-3 is a shared device, so that the size of the detection terminal can be reduced better.
In this embodiment, the infrared wavefront sensing unit S2 and the wide-angle lens unit S2-3 together form an infrared wavefront sensing optical path system.
The wide-angle lens unit S2-3 of this embodiment has lens L1~ L3, diaphragm L4, lens L5~ L9 and beam splitter L10 that set gradually, and infrared light signal and visible light imaging signal can reach beam splitter L10 after L1-L9 in proper order, and beam splitter L10 department forms transmission light path and reflection light path, and then realizes that infrared light signal and visible light imaging signal arrive infrared wavefront sensing unit S2 and visible light imaging unit S3 department respectively.
In the embodiment, antireflection films for antireflection of infrared light signals (1550 nm wavelength) and visible light imaging signals can be arranged on the mirror surfaces of the lenses L1-L3 and the lenses L5-L9, and the transmittance of the antireflection films is designed to be not lower than 99%; the incident mirror surface of the beam splitter L10 is provided with a reflection increasing film (reflectivity about 99%) for increasing reflection of visible light imaging signals and an anti-reflection film (transmittance not less than 99.8%) for increasing reflection of infrared light signals (1550 nm wavelength), so that formation of a transmission light path and a reflection light path can be preferably realized.
As a specific example in this embodiment, the parameters of the respective components of the wide angle lens unit S2-3 are shown in Table 1;
TABLE 1 parameters of the various components of Wide Angle lens Unit S2-3
In the above table, the lenses L1, L5, L7-L10, etc. are all single lenses, and therefore all have only an incident mirror surface and an exit mirror surface; the lenses L2, L3, L6 and the like are formed by compounding and adhering two lenses, so that the lens has an incident mirror surface, a middle mirror surface and an emergent mirror surface; the diaphragm L4 has no thickness and thus has only a diaphragm surface.
In this embodiment, the infrared wavefront sensing unit S2 has an aplanatic collimator lens L12, an infrared narrow-band filter L13, a microlens array L14, and an infrared CCD image sensor, which are sequentially arranged. The acquisition of the spot information can be preferably achieved.
In this embodiment, the parameters of the aplanatic collimator lens L12 can be designed as follows: the diameter is 25.4mm, the focal length is 35.0 mm, the center thickness is 13.3 mm, the edge thickness is 8.8 mm, the curvature radius of an incident mirror surface is 23.2 mm, and the curvature radius of an emergent mirror surface is-17.9 mm; meanwhile, an antireflection film capable of increasing the transmittance of infrared light signals with the wavelength of 1100-1650 nm can be plated at the position of the aplanatic collimating lens L12, and the aplanatic collimating lens L12 can be arranged in a standard lens sleeve and is connected with an infrared CCD (charge coupled device) image sensor through an external sleeve.
In this embodiment, the infrared narrow-band filter L13 has a wavelength of 1550nm, so that interference of other stray light can be prevented.
In this embodiment, the parameters of the microlens array L14 can be designed as follows: the total aperture is 25.4mm, the aperture of the usable light-passing aperture is 15mm multiplied by 15mm, the material is N-BK7 glass, the size of the single micro-lens is 300 mu m multiplied by 300 mu m, and the focal length is 50mm.
In this embodiment, an infrared light signal generated at the beacon light source S1 can be compressed in view angle through the wide-angle lens unit S2-3, and reaches the aplanatic collimator L12 through the transmission light path at the beam splitter L10, and the corresponding light signal can be corrected to approximate collimated light through the aplanatic collimator L12, and then stray light can be filtered out through the infrared narrow-band filter L13, and then the corresponding light signal can be divided through the microlens array L14 to form a plurality of light spots, and then the acquisition of light spot information can be realized through the infrared CCD image sensor.
As shown in fig. 4, the visible light imaging unit S3 and the wide-angle lens unit S2-3 together constitute a visible light imaging optical path system, and the visible light imaging unit S3 includes a visible light CCD image sensor. Therefore, a plane image of the working area of the lifting hook on the vertical plane can be obtained better.
In this embodiment, the visible CCD image sensor can be equipped with an autofocus assembly, so that better resolution can be obtained.
In this embodiment, a convolutional neural network model is constructed at the data processing unit S4, and the convolutional neural network model is used for inputting light spot information and outputting space coordinate information of the beacon light source S1. Therefore, compared with the traditional centroid algorithm, the centroid algorithm and the like, the embodiment has higher response speed.
In this embodiment, the data processing unit S4 can be implemented based on an upper computer such as a computer, and in actual operation, the data processing unit S4 can be connected to the infrared CCD image sensor and the visible CCD image sensor through data lines, so that the light spot information and the plane image can be acquired in time.
In this embodiment, the convolutional neural network model can adopt a modified VGG Net five-convolution group structure, and can complete training in simulation environments such as Matlab and Zemax.
Referring to fig. 5, the convolutional neural network model of the present embodiment has five convolutional groups connected in sequence, and the number of layers is designed to be [3,3,5,5,5], respectively. When the convolutional neural network model is trained, each light spot information and the corresponding spatial coordinate information can be used as a group of samples. In this embodiment, 20000 samples can be constructed as a training set, and 5000 samples are randomly extracted from the training set as a test set. And the optimization target of training can be designed to be that the misjudgment rate is less than 0.2%. The misjudgment rate can be defined as the probability that the distance between the three-dimensional coordinate output by the convolutional neural network model and the real three-dimensional coordinate is less than 1 cm.
Referring to fig. 6, samples can be obtained by simulation using ZEMAX software. Specifically, the optical elements can be modeled on the beacon light source S1, the infrared wavefront sensing unit S2 and the visible light imaging unit S3 in the ZEMAX software, and the corresponding spot information can be preferably acquired by setting the position (i.e., three-dimensional coordinates) of the beacon light source S1 for multiple times, so that the acquisition of the sample can be preferably realized.
By the above, the spatial coordinate information of the beacon light source S1 can be preferably acquired based on the spot information.
In the embodiment, a coordinate conversion model is constructed at the data processing unit S4, and the coordinate conversion model is used for inputting the two-dimensional coordinate of the beacon light source S1 on the vertical surface and outputting the two-dimensional coordinate of the hoisted object on the vertical surface; in particular to a method for preparing a high-performance nano-silver alloy,
wherein,
and
the coordinates of the beacon light source S1 in the horizontal direction and the vertical direction on the vertical plane,
and
respectively are the coordinates of the hoisted object in the horizontal direction and the vertical direction on the vertical surface,
and
the coordinates of the beacon light source S1 in the horizontal direction and the vertical direction in the plane image,
and
respectively the coordinates of the hoisted object in the horizontal direction and the vertical direction in the plane image "
"is the product operation.
By the method, the two-dimensional coordinate of the hoisted object can be acquired based on the space coordinate information of the position (the lifting hook position) of the beacon light source S1.
See fig. 7, which is a schematic illustration of a planar image at the hook working area. Wherein,
the point is where the beacon light source S1 is located,
the point is the edge position of the hoisting object; it will be appreciated that the data processing unit S4 is capable of implementing the location of the beacon light source S1 based on an image recognition algorithm
Point and edge position of hoisted object
And (6) acquiring points. Meanwhile, in the embodiment of the present invention,
the coordinates of the points are (
，
），
The coordinates of the points are (
，
) Which are all pixel coordinates in the planar image.
Fig. 8 is a schematic view of the imaging of the planar image. It can be known that there is a corresponding relationship between the point a where the beacon light source S1 is actually located and the point B where the edge of the object is actually located, and therefore the actual coordinates based on the point a (
，
) The actual coordinates of the point B can be obtained based on the coordinate transformation model (preferably
，
）。
Based on the above, the acquisition of the two-dimensional coordinate of the edge of the lifting object on the vertical plane can be better realized based on the spatial coordinate information of the position of the beacon light source S1.
In this embodiment, the data processing unit S4 is configured to generate and output an alarm signal when any one of the coordinates of the hoisted object in the horizontal direction and the vertical direction on the vertical surface exceeds the coordinate of the obstacle in the corresponding direction on the vertical surface. Therefore, the lifting hook and the lifting object can be better monitored and alarmed.
Based on the system, the embodiment further provides a wavefront sensing and imaging combined type three-dimensional tracking method for hook positioning, which comprises the following steps:
a beacon light source S1 is arranged at the position of the hook, and the beacon light source S1 is used for generating an infrared light signal;
acquiring light spot information of the infrared light signal through an infrared wavefront sensing unit S2, wherein the light spot information is an array light spot of the infrared light signal after being segmented;
acquiring a plane image at the working area of the lifting hook through a visible light imaging unit S3, wherein the imaging surface of the plane image is a vertical surface;
the method comprises the steps that space coordinate information of a beacon light source S1 is obtained through a data processing unit S4 based on light spot information, two-dimensional coordinates of a hoisting object at a lifting hook on a vertical surface are obtained based on the space coordinate information of the beacon light source S1 and a plane image, and alarm information is obtained based on the two-dimensional coordinates of an obstacle at a working area of the lifting hook on the vertical surface and the two-dimensional coordinates of the hoisting object on the vertical surface.
In this embodiment, through the wavefront sensing and imaging combined type three-dimensional tracking system, the actual three-dimensional coordinate of the beacon light source S1 can be directly recovered from the light spot information, the actual two-dimensional position of the hoisted object can be directly calculated and obtained according to the plane image, and the cross-border alarm detection of the lifting hook and the hoisted object can be preferably realized by comparing the actual two-dimensional position with the two-dimensional position of the obstacle. The method can better avoid calibration work, has simple and convenient integral structure and convenient deployment, and can be better suitable for hoisting operation of electric power construction.
As shown in fig. 9 and 10, in practical use, the number and the direction of the detection terminals need to be deployed according to the distribution situation of the obstacles at the working area of the lifting hook.
It can be understood that the final out-of-range determination to be achieved by the method of the present embodiment is performed in the imaging plane. Therefore, the number of the required imaging surfaces needs to be determined based on the obstacle distribution on site, and it can be understood that each imaging surface is a vertical surface in which the largest obstacle information including the height direction and the directions on both sides is required.
In the embodiment, only 1 detection terminal is deployed as an example for explanation, as shown in fig. 9, an electric wire is arranged above an imaging plane, and the electric wire defines the heights of a lifting hook and a lifting object; as can be seen in fig. 10, the imaging plane has poles on both sides, which define the swing width of the hook and the sling.
It will be appreciated that the spatial position information of the obstacle is typically a known quantity or can be obtained directly by conventional measurement means. Therefore, subsequent border-crossing monitoring and alarming can be preferably realized.
In actual deployment, the beacon light source S1 can be provided at the hook or boom tip. The detection terminal can be arranged at the ground end, and the existence of the field angle of the detection terminal is considered, so that the obstacle needs to be ensured to be completely positioned in the imaging plane, and the minimum distance d between the detection terminal and the beacon light source S1 (the horizontal distance between the detection terminal and the crane can be approximately adopted) needs to be arranged.
It can be understood that, in order to ensure that the obstacle is always located in the imaging plane, the measurable height H of the detection terminal needs to be larger than the maximum height H of the obstacle, and the measurable widthlA maximum spacing L greater than the obstacle is required. Wherein, the measurable height h and the measurable widthlThe relationship with the half field angle θ is: h = d × tan θ,l= 2d×tanθ。
thus, the spacing d needs to satisfy: d > H/(d × tan θ) and d > L/2d × tan θ. In this embodiment, θ is designed to be 50 °, so the distance d needs to satisfy: d >0.84H and d > 0.42L.
In the practical use process of the method, the detection terminal can acquire the corresponding plane image and the corresponding light spot information in real time, and the corresponding plane image and the corresponding light spot information can be transmitted to the data processing unit S4 in real time, so that the practical two-dimensional coordinates of the lifting hook and the lifting object can be better realized, and the practical two-dimensional coordinates of the lifting hook and the lifting object can be compared with the practical two-dimensional coordinates of the obstacle, so that the cross-border monitoring and the alarm of the lifting hook and the lifting object can be realized.
It can be understood that, for the obstacles in linear distribution, the model establishment on the vertical plane (i.e. the image plane) can be realized by actually measuring the actual space coordinates of a plurality of point locations; therefore, after actual two-dimensional coordinates of the lifting hook and the lifting object are obtained, the distance between the current positions of the lifting hook and the lifting object and the barrier can be preferably obtained based on the distance calculation between the point lines, and by setting an alarm threshold (such as 1 meter), an alarm signal can be preferably generated in time when the distance between the current positions of the lifting hook and the lifting object and the barrier is lower than the alarm threshold, so that out-of-range monitoring and alarming of power construction lifting operation can be preferably realized.
It is understood that for irregularly distributed obstacles, the constraints on the obstacles can be achieved by establishing a set of two-dimensional coordinates of all points or key points of the obstacles at the vertical plane (i.e., the image plane); therefore, after the actual two-dimensional coordinates of the lifting hook and the lifting object are obtained, the distance between the current positions of the lifting hook and the lifting object and the obstacle can be preferably obtained based on the calculation of the distance between points one by one, and by setting an alarm threshold (such as 1 meter), an alarm signal can be preferably generated in time when the minimum distance between the current positions of the lifting hook and the lifting object and the obstacle is lower than the alarm threshold, so that the border-crossing monitoring and alarm of the hoisting operation of the power construction can be preferably realized.
It is easily understood that a person skilled in the art can combine, split, recombine and the like the embodiments of the present application to obtain other embodiments on the basis of one or more embodiments provided by the present application, and the embodiments do not go beyond the protection scope of the present application.
The present invention and the embodiments thereof have been described above schematically, the description is not restrictive, the embodiments shown in the examples are only part of the embodiments of the present invention, and the actual structure is not limited thereto. Therefore, without departing from the spirit of the present invention, a person of ordinary skill in the art should understand that the present invention shall not be limited to the embodiments and the similar structural modes without creative design.
Claims (10)
1. Wavefront sensing and imaging combined type three-dimensional tracking system for hook positioning comprises:
a beacon light source (S1) for being provided at the hook and for generating an infrared light signal;
the infrared wavefront sensing unit (S2) is used for acquiring light spot information of the infrared light signal, and the light spot information is array light spots obtained after the infrared light signal is subjected to segmentation processing;
the visible light imaging unit (S3) is used for acquiring a plane image at the working area of the lifting hook, and the imaging surface of the visible light imaging unit is a vertical surface; and the number of the first and second groups,
and the data processing unit (S4) is used for acquiring the space coordinate information of the beacon light source (S1) based on the light spot information, acquiring the two-dimensional coordinate of the hoisting object at the lifting hook on the vertical surface based on the space coordinate information of the beacon light source (S1) and the plane image, and acquiring alarm information based on the two-dimensional coordinate of the obstacle at the working area of the lifting hook on the vertical surface and the two-dimensional coordinate of the hoisting object on the vertical surface.
2. The wavefront sensing and imaging combined three-dimensional tracking system for hook positioning according to claim 1, wherein: the infrared wavefront sensing unit (S2) and the visible light imaging unit (S3) share the same wide-angle lens unit (S2-3), and the wide-angle lens unit (S2-3) is provided with a transmission light path and a reflection light path; the transmission optical path is used for transmitting the infrared light signal to the infrared wavefront sensing unit (S2), and the reflection optical path is used for transmitting the visible light imaging signal to the visible light imaging unit (S3).
3. A hook positioning wavefront sensing and imaging hybrid three-dimensional tracking system as claimed in claim 1 or 2, wherein: the infrared wavefront sensing unit (S2) is provided with an aplanatic collimating lens (L12), an infrared narrow-band filter (L13), a micro-lens array (L14) and an infrared CCD image sensor which are arranged in sequence.
4. A hook positioning wavefront sensing and imaging hybrid three-dimensional tracking system as claimed in claim 1 or 2, wherein: the visible light imaging unit (S3) includes a visible light CCD image sensor.
5. The hook positioning wavefront sensing and imaging hybrid three-dimensional tracking system of claim 1, further comprising: and a convolutional neural network model is constructed at the data processing unit (S4), and the convolutional neural network model is used for inputting light spot information and outputting space coordinate information of the beacon light source (S1).
6. The hook positioning wavefront sensing and imaging hybrid three-dimensional tracking system of claim 1, further comprising: a coordinate conversion model is constructed at the data processing unit (S4), and the coordinate conversion model is used for inputting the two-dimensional coordinates of the beacon light source (S1) on the vertical surface and outputting the two-dimensional coordinates of the hoisted object on the vertical surface; in particular to a method for preparing a high-performance nano-silver alloy,
wherein,
and
coordinates of the beacon light source (S1) in the horizontal direction and the vertical direction on the vertical plane,
and
respectively are the coordinates of the hoisted object in the horizontal direction and the vertical direction on the vertical surface,
and
coordinates of the beacon light source (S1) in the horizontal direction and the vertical direction in the plane image,
and
respectively the coordinates of the hoisted object in the horizontal direction and the vertical direction in the plane image "
"is the product operation.
7. The hook positioning wavefront sensing and imaging hybrid three-dimensional tracking system of claim 1, further comprising: the data processing unit (S4) is used for generating and outputting an alarm signal when any one of the coordinates of the hoisted object in the horizontal direction and the vertical direction on the vertical surface exceeds the coordinate of the obstacle in the corresponding direction on the vertical surface.
8. The wave-front sensing and imaging combined three-dimensional tracking method for hook positioning comprises the following steps:
arranging a beacon light source (S1) at the hook, wherein the beacon light source (S1) is used for generating an infrared light signal;
acquiring light spot information of the infrared light signal through an infrared wavefront sensing unit (S2), wherein the light spot information is an array light spot of the infrared light signal after being segmented;
acquiring a plane image at the working area of the lifting hook by a visible light imaging unit (S3), wherein the imaging surface of the plane image is a vertical surface;
the method comprises the steps that space coordinate information of a beacon light source (S1) is obtained through a data processing unit (S4) based on light spot information, a two-dimensional coordinate of a hoisting object at a lifting hook on a vertical surface is obtained based on the space coordinate information of the beacon light source (S1) and a plane image, and alarm information is obtained based on the two-dimensional coordinate of an obstacle at a lifting hook working area on the vertical surface and the two-dimensional coordinate of the hoisting object on the vertical surface.
9. The method of claim 8, wherein the wavefront sensing and imaging compound three-dimensional tracking method for hook positioning comprises: and constructing a convolutional neural network model at the data processing unit (S4), wherein the convolutional neural network model is used for inputting the light spot information and outputting the space coordinate information of the beacon light source (S1).
10. The method of claim 8, wherein the wavefront sensing and imaging compound three-dimensional tracking method for hook positioning comprises: a coordinate conversion model is constructed at the data processing unit (S4), and the coordinate conversion model is used for inputting the two-dimensional coordinates of the beacon light source (S1) on the vertical surface and outputting the two-dimensional coordinates of the hoisted object on the vertical surface; in particular to a method for preparing a high-performance nano-silver alloy,
wherein,
and
coordinates of the beacon light source (S1) in the horizontal direction and the vertical direction on the vertical plane,
and
respectively are the coordinates of the hoisted object in the horizontal direction and the vertical direction on the vertical surface,
and
coordinates of the beacon light source (S1) in the horizontal direction and the vertical direction in the plane image,
and
respectively water of the hoisted objects in the plane imageCoordinates in the horizontal and vertical directions'
"is the product operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310058033.0A CN115784026B (en) | 2023-01-18 | 2023-01-18 | Wavefront sensing and imaging combined type three-dimensional tracking system and method for positioning lifting hook |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310058033.0A CN115784026B (en) | 2023-01-18 | 2023-01-18 | Wavefront sensing and imaging combined type three-dimensional tracking system and method for positioning lifting hook |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115784026A true CN115784026A (en) | 2023-03-14 |
| CN115784026B CN115784026B (en) | 2023-04-07 |
Family
ID=85429728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310058033.0A Active CN115784026B (en) | 2023-01-18 | 2023-01-18 | Wavefront sensing and imaging combined type three-dimensional tracking system and method for positioning lifting hook |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115784026B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101738721A (en) * | 2009-12-25 | 2010-06-16 | 中国科学院光电技术研究所 | Adaptive optical imaging system of Hartmann wavefront sensor without system error |
| CN108291953A (en) * | 2015-12-23 | 2018-07-17 | 阿森蒂亚影像有限公司 | Positioning system and correlation technique |
| US20190224846A1 (en) * | 2016-07-15 | 2019-07-25 | Fastbrick Ip Pty Ltd | Dynamic compensation of a robot arm mounted on a flexble arm |
| CN111017726A (en) * | 2019-11-19 | 2020-04-17 | 中联重科股份有限公司 | Crane hook positioning method, device and system and engineering machinery |
| US20200307661A1 (en) * | 2016-10-20 | 2020-10-01 | Rail Vision Ltd | System and method for object and obstacle detection and classification in collision avoidance of railway applications |
| CN112504177A (en) * | 2020-12-01 | 2021-03-16 | 中国人民解放军国防科技大学 | Multifunctional vertical zero-position overlapping scanning interference measuring device |
| CN114644291A (en) * | 2022-05-23 | 2022-06-21 | 安徽送变电工程有限公司 | Safety boundary building device for electric power hoisting operation based on visual processing |
| CN114967165A (en) * | 2022-05-19 | 2022-08-30 | 中国人民解放军国防科技大学 | Wavefront sensing system adopting dual-channel polarization coding and wavefront information decoding method |
-
2023
- 2023-01-18 CN CN202310058033.0A patent/CN115784026B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101738721A (en) * | 2009-12-25 | 2010-06-16 | 中国科学院光电技术研究所 | Adaptive optical imaging system of Hartmann wavefront sensor without system error |
| CN108291953A (en) * | 2015-12-23 | 2018-07-17 | 阿森蒂亚影像有限公司 | Positioning system and correlation technique |
| US20190224846A1 (en) * | 2016-07-15 | 2019-07-25 | Fastbrick Ip Pty Ltd | Dynamic compensation of a robot arm mounted on a flexble arm |
| US20200307661A1 (en) * | 2016-10-20 | 2020-10-01 | Rail Vision Ltd | System and method for object and obstacle detection and classification in collision avoidance of railway applications |
| CN111017726A (en) * | 2019-11-19 | 2020-04-17 | 中联重科股份有限公司 | Crane hook positioning method, device and system and engineering machinery |
| CN112504177A (en) * | 2020-12-01 | 2021-03-16 | 中国人民解放军国防科技大学 | Multifunctional vertical zero-position overlapping scanning interference measuring device |
| CN114967165A (en) * | 2022-05-19 | 2022-08-30 | 中国人民解放军国防科技大学 | Wavefront sensing system adopting dual-channel polarization coding and wavefront information decoding method |
| CN114644291A (en) * | 2022-05-23 | 2022-06-21 | 安徽送变电工程有限公司 | Safety boundary building device for electric power hoisting operation based on visual processing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115784026B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109544679B (en) | Three-dimensional reconstruction method for inner wall of pipeline | |
| US10412283B2 (en) | Dual aperture 3D camera and method using differing aperture areas | |
| CN109375237A (en) | A kind of all solid state face array three-dimensional imaging laser radar system | |
| CN104091168B (en) | Power line automatic extraction and positioning method based on unmanned aerial vehicle image | |
| CN112969937A (en) | LIDAR system and method | |
| KR20180132809A (en) | A detector for optically detecting one or more objects | |
| CN110132226B (en) | System and method for measuring distance and azimuth angle of unmanned aerial vehicle line patrol | |
| CN207380238U (en) | A kind of laser radar based on diffraction optics | |
| KR20160001906A (en) | Method and apparatus for implemeting active imaging system | |
| US12008783B2 (en) | Reality capture device | |
| CN111175786A (en) | Multi-path crosstalk-eliminating wide-view-field high-resolution solid-state laser radar | |
| CN114690796A (en) | Relative positioning system and method for distance measurement between unmanned aerial vehicles | |
| JP7541035B2 (en) | Scanning lidar with optical switching | |
| CN108646239A (en) | A kind of detection and identification device based on radar and photoelectric technology | |
| CN114008483A (en) | System and method for time-of-flight optical sensing | |
| CN113785217A (en) | Electro-optical system and method for scanning illumination onto a field of view | |
| CN109901191A (en) | Rotary scanning type laser infrared radar imaging device | |
| CN112269181A (en) | Laser active detection device and laser active detection processing system | |
| CN114112972B (en) | Closed space SF 6 Gas leakage infrared telemetry device and imaging positioning method thereof | |
| CN115784026B (en) | Wavefront sensing and imaging combined type three-dimensional tracking system and method for positioning lifting hook | |
| CN211291443U (en) | Passive telemetering robot based on live-action three-dimensional map | |
| CN209624768U (en) | laser active detection device | |
| Zheng et al. | Time-sensitive and distance-tolerant deep learning-based vehicle detection using high-resolution radar bird's-eye-view images | |
| Li et al. | Safety monitoring method for powerline corridors based on single‐stage detector and visual matching | |
| CN116692690A (en) | Crane anti-collision early warning method, device, equipment and medium |
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 |