CN109883391B - Monocular distance measurement method based on digital imaging of microlens array - Google Patents
Monocular distance measurement method based on digital imaging of microlens array Download PDFInfo
- Publication number
- CN109883391B CN109883391B CN201910210286.9A CN201910210286A CN109883391B CN 109883391 B CN109883391 B CN 109883391B CN 201910210286 A CN201910210286 A CN 201910210286A CN 109883391 B CN109883391 B CN 109883391B
- Authority
- CN
- China
- Prior art keywords
- image
- refocusing
- digital
- sample
- target
- 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
Landscapes
- Studio Devices (AREA)
Abstract
The invention relates to a monocular distance measurement method based on digital imaging of a micro-lens array, which comprises the following steps: acquiring a sample four-dimensional light field image of a reference sample by adopting a micro-lens array camera, and carrying out digital refocusing on the sample four-dimensional light field image through different digital refocusing parameters to obtain a plurality of sample refocusing images; obtaining the distance measurement distance of the clearest partition corresponding to the digital imaging focal plane from the sample refocusing image; fitting different digital refocusing parameters and corresponding distance measurement distances to obtain a distance measurement fitting curve; then, a micro-lens array camera is adopted to obtain a target four-dimensional light field image, and digital refocusing parameters are selected to carry out digital refocusing on the target four-dimensional light field image to obtain a target refocusing image; and processing the refocused image of the target, and combining a distance measurement fitting curve to obtain the target distance. The invention can obtain image sequences of different depths only through an imaging algorithm after single exposure, and can be used for presenting dynamic change and measuring the distance of a target with high real-time requirement.
Description
Technical Field
The invention relates to the technical field of digital imaging ranging, in particular to a monocular ranging method based on digital imaging of a micro-lens array.
Background
The micro-lens array imaging is a novel imaging technology, and light rays in different directions penetrate through the pupil of the main lens and then penetrate through the micro-lens by installing a micro-lens array between the main lens and the sensor to form an image on an image detector; light of different directions impinges on different pixels, so that the pixels on the detector are able to record four-dimensional light field data containing direction information and intensity information of the light rays. And the four-dimensional light field data is subjected to later digital focusing, so that the depth of a scene in a larger range can be effectively captured, and a real three-dimensional scene is displayed.
At present, the photoelectric distance measuring technology is increasingly required in the fields of unmanned aircraft navigation, satellite docking, particle image rapid concentration measurement, industrial flame detection and the like. The existing photoelectric distance measuring method mainly comprises multi-eye distance measuring and monocular distance measuring; the multi-view distance measurement mainly comprises the steps of acquiring projections of a target from different directions by using a plurality of cameras, and calculating the coordinate distance of the target by using stereo matching and a visual algorithm. The multi-view distance measurement method has the problems of multiple devices, complex optical path, long time consumption, difficulty in synchronizing camera signals and the like. Monocular distance measurement obtains the collected images of the target on different focal planes by changing the mechanical parameters of the camera or changing the position of the camera, and then obtains the coordinate position of the target by changing the definition of the image. When the monocular distance measurement method changes the focusing surface of the imaging system, a process of focus moving or mechanical zooming is required, and a delay error is generated, so that the monocular distance measurement method cannot be applied to the measurement field with higher real-time requirement.
Therefore, in order to overcome the above disadvantages, it is necessary to provide a monocular distance measurement method, so that it only needs to perform a single exposure on the target to calculate and obtain image sequences of different depths, thereby satisfying the real-time requirement of distance measurement.
Disclosure of Invention
The invention aims to solve the technical problem of providing a monocular distance measuring method based on digital imaging of a micro-lens array, aiming at the defect that the existing monocular distance measuring method needs to move focus or mechanically zoom when the focusing surface of an imaging system is changed, so that the distance measuring generates a delay error.
In order to solve the technical problem, the invention provides a monocular distance measurement method based on microlens array digital imaging, which comprises the following steps:
acquiring a sample four-dimensional light field image of a reference sample by adopting a micro-lens array camera, and carrying out digital refocusing on the sample four-dimensional light field image through different digital refocusing parameters to obtain a plurality of sample refocusing images;
performing definition calculation on different partitions of each sample refocused image to determine the clearest partition, taking the position corresponding to the clearest partition as a digital imaging focus plane under the current digital refocusing parameter, and calculating the distance measurement distance of the clearest partition by using the digital imaging focus plane;
calculating different digital imaging focal planes corresponding to different digital refocusing parameters, and determining the distance measurement distance of the clearest partition in the refocusing images of different samples; fitting different digital refocusing parameters and corresponding distance measurement distances to obtain a distance measurement fitting curve;
then, a micro-lens array camera is adopted to obtain a target four-dimensional light field image, and digital refocusing parameters are selected to carry out digital refocusing on the target four-dimensional light field image to obtain a target refocusing image;
and processing the refocused image of the target, and combining a distance measurement fitting curve to obtain the target distance.
In the monocular distance measuring method based on digital imaging of the microlens array according to the present invention, the method of obtaining the target distance by combining the distance measuring fitting curve includes:
selecting a plurality of digital refocusing parameters to obtain a plurality of target refocusing images, performing definition calculation on the plurality of target refocusing images, and obtaining target distances by combining a distance measurement fitting curve based on the digital refocusing parameters corresponding to the clearest target refocusing image;
or selecting a digital refocusing parameter to obtain a target refocusing image, performing definition calculation on different partitions of the target refocusing image, and determining the clearest partition; and then obtaining the target distance corresponding to the clearest partition based on the digital refocusing parameters and a distance measurement fitting curve.
In the monocular distance measuring method based on the digital imaging of the microlens array, the reference sample comprises a graduated scale, and the graduated scale is obliquely arranged on a slide rail; the inclined plane of the graduated scale corresponds to the lens of the micro-lens array camera.
In the monocular distance measuring method based on digital imaging of a microlens array according to the present invention, the calculating the sharpness of the different partitions of each sample refocused image includes:
and calculating a digital imaging focal plane corresponding to the clearest partition in each sample refocusing image by using a digital focusing algorithm.
In the monocular distance measuring method based on digital imaging of a microlens array according to the present invention, the calculating the sharpness of the different partitions of each sample refocused image further includes:
and performing definition calculation on all the partitions in each sample refocusing image by using a definition evaluation function to obtain a definition curve of all the partitions in each sample refocusing image.
In the monocular distance measuring method based on digital imaging of a microlens array according to the present invention, the method of obtaining the distance measuring distance of the clearest partition includes:
and determining the clearest partition according to the definition curve of each sample refocusing image, and calculating to obtain the distance measurement distance corresponding to the current digital refocusing parameter according to the digital imaging focus plane corresponding to the clearest partition.
In the monocular distance measuring method based on microlens array digital imaging according to the present invention, the partition method of different partitions of the sample refocusing image comprises:
the tilt directions corresponding to the sample refocused image are equally divided into a plurality of partitions.
In the monocular distance measuring method based on digital imaging of the microlens array according to the present invention, the method for obtaining the digital imaging focal plane includes:
according to the formula of the digital focusing algorithm, calculating to obtain a digital imaging focal plane image E corresponding to the clearest partition in the sample refocusing image corresponding to the digital refocusing parameter alphaαF(x,y)：
In the formula LαF(u, v, x, y) is the spectral energy carried by the point (x, y) on the x-y plane through which the light ray on the F image plane passes from the point (u, v) on the u-v plane at the image distance l ═ F; the u-v plane represents the plane of the micro lens array, and the x-y plane represents the plane of the detector; l isFRepresenting the spectral energy carried by a projection of a point (u, v) on the u-v plane onto a point (x, y) on the x-y plane after entering the microlens array camera.
In the monocular distance measuring method based on digital imaging of the microlens array according to the present invention, the method of performing sharpness calculation for all the regions in each sample refocused image using a sharpness evaluation function includes:
the sharpness evaluation result f (i) for each partition was calculated using the following formula:
wherein C is a gradient matrix.
In the monocular distance measuring method based on microlens array digital imaging according to the present invention, the gradient matrix C is:
C＝I*L，
in the formula, I is the gray value corresponding to the pixels of different partitions in the sample refocusing image, and the digital imaging focal plane image E corresponding to different partitionsαF(x, y); the L is a Laplacian operator;
when the sample refocused image of the scale is divided equally into 60 sections along the scale direction:
the monocular distance measuring method based on the digital imaging of the microlens array has the following beneficial effects: the invention is provided based on the characteristic that the micro-lens array digital imaging technology takes pictures first and then focuses. Photographing a reference sample by using a micro-lens array camera to obtain a four-dimensional light field image of the sample; refocusing the sample four-dimensional light field image by using a mathematical focusing algorithm to obtain refocused images under different depths of field; finally, judging the focusing position of the refocused image of the sample under different depths of field through definition calculation, thereby determining the distance measurement distance of the sample; and obtaining a fitting curve for calibration through multiple times of calculation, and obtaining the distance measurement distance of the target four-dimensional light field image by adopting the fitting graph.
The method of the invention obtains refocused images under different focusing depths after carrying out digital focusing processing on the originally obtained sample four-dimensional light field image, and the effect of shooting images of different depths after the mechanical zooming of the traditional camera is the same. However, the digital focusing imaging does not have a mechanical zooming process, and image sequences with different depths can be obtained only through an imaging algorithm after a single exposure. Therefore, the method of the invention has simple system light path and can be used for presenting dynamic change and measuring the distance of the target with high real-time requirement.
Drawings
FIG. 1 is a schematic diagram of exemplary parametric characterization of a monocular distance measuring method based on microlens array digital imaging according to the present invention; in the figure, A represents the plane of the micro-lens array, and B represents the plane of the detector;
FIG. 2 is a schematic view of a microlens array camera imaging according to the present invention; in the figure, C denotes the object plane and F denotes the main lens;
FIG. 3 is a schematic view of digital imaging focusing according to the present invention; in the figure, E represents a digital imaging focal plane;
FIG. 4 is an exemplary schematic diagram of distance calibration using a scale; in the figure, 1 denotes a slide rail, 2 denotes a microlens array camera, and 3 denotes a scale;
fig. 5 is a schematic diagram of a scale refocused image obtained when α is 5.97;
fig. 6 is a schematic diagram of a scale refocused image obtained when α is 6.73;
fig. 7 is a schematic diagram of a scale refocused image obtained when α is 7.81;
fig. 8 is a sharpness curve of the refocused image of the scale when α is 6.73;
fig. 9 is a graph showing the correspondence between the digital refocusing parameter α and the actual distance D of the reference sample.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
In a first embodiment, the present invention provides a monocular distance measurement method based on digital imaging of a microlens array, which is shown in fig. 1 to 4, and includes:
acquiring a sample four-dimensional light field image of a reference sample by adopting a micro-lens array camera, and carrying out digital refocusing on the sample four-dimensional light field image through different digital refocusing parameters to obtain a plurality of sample refocusing images;
performing definition calculation on different partitions of each sample refocused image to determine the clearest partition, taking the position corresponding to the clearest partition as a digital imaging focus plane under the current digital refocusing parameter, and calculating the distance measurement distance of the clearest partition by using the digital imaging focus plane;
calculating different digital imaging focal planes corresponding to different digital refocusing parameters, and determining the distance measurement distance of the clearest partition in the refocusing images of different samples; fitting different digital refocusing parameters and corresponding distance measurement distances to obtain a distance measurement fitting curve;
then, a micro-lens array camera is adopted to obtain a target four-dimensional light field image, and digital refocusing parameters are selected to carry out digital refocusing on the target four-dimensional light field image to obtain a target refocusing image;
and processing the refocused image of the target, and combining a distance measurement fitting curve to obtain the target distance.
In the embodiment, the calibration of the digital refocusing parameter and the ranging distance is realized through data acquired by a reference sample. The target distance is then determined using the fitted curve in actual ranging use. The micro-lens array camera can obtain four-dimensional light field image data of a reference sample through single exposure.
After the sample four-dimensional light field image is obtained, the image under different depths of field can be obtained by refocusing the sample four-dimensional light field image by using a digital focusing algorithm. And judging the focusing position of the image under different depths of field by utilizing the definition evaluation function, and corresponding the position to the distance value between the reference sample and the camera, thereby completing the distance measurement.
The reference sample may be a graduated scale to facilitate distance conversion.
Further, the following method may be adopted for ranging of a single target:
the method for obtaining the target distance by combining the ranging fitting curve comprises the following steps:
selecting a plurality of digital refocusing parameters to obtain a plurality of target refocusing images, calculating the definition of the plurality of target refocusing images, and obtaining the target distance by combining a distance measurement fitting curve based on the digital refocusing parameter corresponding to the clearest target refocusing image. In this case, the target distance is measured regardless of the size of the target.
Or still further, the following method may be adopted for ranging of multiple targets or multiple partitions in the image:
the method for obtaining the target distance by combining the ranging fitting curve comprises the following steps:
selecting a digital refocusing parameter to obtain a target refocusing image, and performing definition calculation on different partitions of the target refocusing image to determine the clearest partition; and then obtaining the target distance corresponding to the clearest partition based on the digital refocusing parameters and a distance measurement fitting curve. In this case, a plurality of targets may exist in the monitoring scene, or the target size is considered to be large, and the corresponding target distance is measured by dividing different partitions.
Further, as shown in connection with fig. 4, the reference sample includes a scale. The scale is selected as a reference sample so as to directly obtain the distance measurement distance of the target.
Still further, as shown in fig. 4, the graduated scale is obliquely placed on the slide rail; the inclined plane of the graduated scale corresponds to the lens of the micro-lens array camera.
The inclination angle that the scale was put can be selected as required.
Still further, as shown in fig. 3, the performing the sharpness calculation on the different regions of each sample refocused image includes:
and calculating a digital imaging focal plane corresponding to the clearest partition in each sample refocusing image by using a digital focusing algorithm. After the digital imaging focal plane image is obtained, the pixel gray value presented by the digital imaging focal plane image is combined, and the digital imaging focal plane image can be used for evaluating subsequent definition.
Still further, as shown in fig. 5 to 8, the calculating the sharpness of the different regions of each sample refocused image further includes:
and performing definition calculation on all the partitions in each sample refocusing image by using a definition evaluation function to obtain a definition curve of all the partitions in each sample refocusing image.
Still further, with reference to fig. 4, the method for obtaining the ranging distance of the clearest partition includes:
and determining the clearest partition according to the definition curve of each sample refocusing image, and calculating to obtain the distance measurement distance corresponding to the current digital refocusing parameter according to the digital imaging focus plane corresponding to the clearest partition.
Still further, with reference to fig. 5 to 7, the method for partitioning different partitions of the sample refocused image includes:
the tilt directions corresponding to the sample refocused image are equally divided into a plurality of partitions.
The refocused image of the graduated scale is equally divided from top to bottom, and can be equally divided into N subareas, wherein the size of N is selected according to needs. For example, the image of 200 × 200 may be divided into ten partitions of 20 × 200, and then the sharpness is calculated for each of the partitions, and the result of the sharpness determination is compared with a larger value, which is the partition where the current focus position is located.
Still further, with reference to fig. 3, the method for obtaining the digital imaging focal plane includes:
according to the formula of the digital focusing algorithm, calculating to obtain a digital imaging focal plane image corresponding to the clearest partition in the sample refocusing image corresponding to the digital refocusing parameter alphaEαF(x,y)：
In the formula LαF(u, v, x, y) is the spectral energy carried by the point (x, y) on the x-y plane through which the light ray on the F image plane passes from the point (u, v) on the u-v plane at the image distance l ═ F; the u-v plane represents the plane of the micro lens array, and the x-y plane represents the plane of the detector; l isFRepresenting the spectral energy carried by a projection of a point (u, v) on the u-v plane onto a point (x, y) on the x-y plane after entering the microlens array camera.
Bottom-up to digital imaging focal plane image EαFThe process of obtaining (x, y) is specifically described:
and (3) light field information characterization:
without taking into account the reasonable assumptions of wavelength change and energy transmission attenuation, the light field can be characterized by two mutually parallel planes (u, v) and (s, t), as shown in fig. 1, where plane (u, v) represents the plane of the microlens array in the microlens array camera and plane (x, y) represents the plane of the detector in the microlens array camera. The light field information refers to the sum of the ray radiance function for each point and each direction in space. The light field information can be characterized by the connection of two points (x, y) and (u, v) in two parallel planes, as shown in fig. 1. 2 coordinate planes are defined in fig. 1: a u-v plane and an x-y plane. With LF(u, v, x, y) represents the spectral energy carried by a ray of light that enters the optical system and is projected through a point (u, v) on the u-v plane to a point (x, y) on the x-y plane, where the point (x, y) receives LFThe incident light flux of (u, v, x, y) is expressed as:
in the formula EF(x, y) is the incident flux of light at coordinates (x, y) on the image plane at the image distance F; f is the image distance; a (u, v) is the area of a detector pixel; theta is a light ray LF(u, v, x, y) is at an angle to the direction of the main optical axis of the system.
Assuming that the microlens array plane and the detector plane are infinite and only light propagating within the u-v plane and x-y plane of the optical system is considered, for ease of analysis, the variation of the cosine of the light and the scale factor 1/F are ignored2. The above formula is simplified to that,
EF(x,y)＝∫∫LF(u,v,x,y)dudv，
therefore, a relatively perfect light field information parameter characterization mode is obtained.
Digital imaging of microlens array:
after the micro-lens array camera takes a picture, light rays in each direction in the four-dimensional light field penetrate through the micro-lens array and then strike different pixels, and as shown in fig. 2, the position, the direction and the intensity information of the light rays are stored in the same original light field image.
The pixel points in the acquired two-dimensional original light field image are subjected to light ray tracing according to a certain rule to correspond to a four-dimensional light field, and the four-dimensional light field is re-projected to a focusing plane with a certain depth for integral superposition to obtain images on different focusing planes, so that the purpose of obtaining images of the focusing planes with multiple depths by single photographing is achieved, and the process of digital focusing imaging is achieved.
Suppose that the imaging system obtains an image E at an image distance l ═ FF(x, y) is unclear, and an image on the image plane at an image distance l ═ α F is a clear image EαF(x, y), α is a coefficient for adjusting the size of the image distance l'. As shown in FIG. 3, the ray passing through the point (x, y) on the F image surface is denoted as LαF(u, v, x, y), the projection coordinate will become L when the light reaches the α F image planeαF. Light LFAnd LαFIs the same light between the lens and the sensor. Suppose a ray LαFThe coordinates of the intersection point of the refocusing plane (x, y), then the coordinates of the intersection point of this ray at the detector plane are (u + (x-u)/α, v + (y-v)/α), LFAnd LαFThe following transformation of coordinates between:
substituting conversion formula into simplified formula EF(x,y)＝∫∫LFIn (u, v, x, y) dudv, an imaging formula of a refocused image plane is obtained:
the digital refocusing parameter α is changed, images on image planes of different focusing planes can be obtained through calculation, but α is only a relative value and cannot represent a real distance in a scene, so that the relationship between α and the real distance needs to be calibrated.
And judging the focusing position according to the definition:
according to the geometrical optics principle, the optical system images an object at a certain distance position, and when the imaging position and the object position meet the conjugate relation, the imaging is clearest at the position, which is also called an ideal image surface. That is to say, the position of the object when the camera forms the clearest image of the object is the actual position of the object, thereby realizing the measurement and calibration of the distance. By utilizing the characteristic, after the image sequence of digital focusing is obtained, the image is distinguished through the definition function, and the clearest position on the image is the focusing position corresponding to the digital refocusing parameter alpha.
Referring to fig. 4, a graduated scale 3 is placed above the slide rail 1 at an included angle, and the reading D on the graduated scale corresponds to the refocusing depth D, i.e. the actual distance D for ranging. Continuously changing the digital refocusing parameter α, performing digital focusing on the sample four-dimensional light field image for multiple times to obtain a series of refocusing images containing depth information and having distinct definitions, as shown in fig. 5 to 7. It can be seen that the upper portion of the refocus image is clear when the focus parameter α is 5.97, the middle portion of the refocus image is clear when the refocus parameter α is 6.73, and the lower portion of the refocus image is clear when the refocus parameter α is 7.81. And equally dividing the refocused image of each scale by N along the scale direction of the refocused image to obtain N subareas, and then calculating the definition of each subarea.
A clear image has clear edge outline and high contrast with the background, namely, the clear position has a larger gradient function value when the gradient evaluation is carried out. Therefore, the method selects a Laplacian definition evaluation function based on the gray gradient to calculate the definition.
Still further, with reference to fig. 8 and 9, a method for performing sharpness calculation on all the partitions in each refocused sample image by using a sharpness evaluation function includes:
the sharpness evaluation result f (i) for each partition was calculated using the following formula:
wherein C is a gradient matrix.
The Laplacian gradient function computes the second order gradient value of the image using the Laplacian operator convolved with the image matrix:
still further, the gradient matrix C is:
C＝I*L，
in the formula, I is the gray value corresponding to the pixels of different partitions in the sample refocusing image, and the digital imaging focal plane image E corresponding to different partitionsαF(x, y); the L is a Laplacian operator;
when the sample refocused image of the scale is divided equally into 60 sections along the scale direction:
after the definition curve diagram is calculated by utilizing the Laplacian definition evaluation function, the part with the maximum definition value is a focusing surface with numerical focusing, and the scale D corresponding to the focusing surface is converted into a horizontal distance D, so that the calibration of the digital focusing distance position can be completed. And (3) carrying out distance position calibration operation on the digital focusing images with different values alpha to obtain a fitting relation curve of the parameter alpha and the actual distance position. So far, the distance position of a certain object in the target scene can be obtained only according to the definition value of the target image and the digital refocusing parameter alpha.
As shown in fig. 8, the sharpness of the image reaches the maximum value when N is 20, and the horizontal distance corresponds to 2.4 cm.
In practical use, the method can realize the ranging process through matlab programming.
The specific embodiment is as follows:
1) the inclined graduated scale with clear scales is photographed by using the micro-lens array camera, the inclination angle can be selected according to the situation, and a four-dimensional light field image of the graduated scale is acquired as shown in fig. 4.
2) Calculating a scale refocusing image under a certain digital refocusing parameter alpha by using a digital focusing algorithm, wherein the formula of the digital focusing algorithm is as follows:
different parameters alpha are selected, refocused images on image surfaces of different focusing planes are obtained through calculation, alpha is only a relative value and cannot represent a real distance position in a scene, and the calibration of the relationship between the parameters alpha and the real distance position can be realized through the following steps.
3) And dividing the refocused image of the scale into N equal parts along the scale direction, wherein N is 60, and calculating the definition of each part by using a definition evaluation function to obtain a definition curve.
C＝I*L，
4) The position with the maximum numerical value on the definition curve diagram is the depth position of numerical value focusing, and the corresponding scale D is converted into the horizontal distance D, so that the calibration of the digital focusing distance position can be completed.
5) And (3) changing the digital refocusing parameter alpha step by step to enable the four-dimensional light field data to be digitally imaged on the focusing surface at different depths, repeating the third step and the fourth step to obtain a plurality of groups of data of the parameter alpha and the distance D, and further determining a corresponding relation fitting curve of the parameter alpha and the actual distance position D. And combining the fitting curve, and acquiring the distance position of a certain target in the scene only according to the image definition value and the digital refocusing parameter alpha after acquiring the target four-dimensional light field image.
In conclusion, the method can obtain image sequences of different depths only by single exposure and calculation of an imaging algorithm, thereby determining the distance measurement distance and being suitable for the real-time distance measurement requirement of a dynamically-changed target.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A monocular distance measurement method based on digital imaging of a micro lens array is characterized by comprising the following steps:
acquiring a sample four-dimensional light field image of a reference sample by adopting a micro-lens array camera, and carrying out digital refocusing on the sample four-dimensional light field image through different digital refocusing parameters to obtain a plurality of sample refocusing images;
performing definition calculation on different partitions of each sample refocused image to determine the clearest partition, taking the position corresponding to the clearest partition as a digital imaging focus plane under the current digital refocusing parameter, and calculating the distance measurement distance of the clearest partition by using the digital imaging focus plane;
calculating different digital imaging focal planes corresponding to different digital refocusing parameters, and determining the distance measurement distance of the clearest partition in the refocusing images of different samples; fitting different digital refocusing parameters and corresponding distance measurement distances to obtain a distance measurement fitting curve;
then, a micro-lens array camera is adopted to obtain a target four-dimensional light field image, and digital refocusing parameters are selected to carry out digital refocusing on the target four-dimensional light field image to obtain a target refocusing image;
processing the target refocusing image, and combining a distance measurement fitting curve to obtain a target distance;
the method for obtaining the target distance by combining the ranging fitting curve comprises the following steps:
selecting a plurality of digital refocusing parameters to obtain a plurality of target refocusing images, performing definition calculation on the plurality of target refocusing images, and obtaining target distances by combining a distance measurement fitting curve based on the digital refocusing parameters corresponding to the clearest target refocusing image;
or selecting a digital refocusing parameter to obtain a target refocusing image, performing definition calculation on different partitions of the target refocusing image, and determining the clearest partition; then, based on the digital refocusing parameters and a distance measurement fitting curve, obtaining a target distance corresponding to the clearest partition;
the reference sample comprises a graduated scale which is obliquely arranged on the slide rail; the inclined plane of the graduated scale corresponds to the lens of the micro-lens array camera.
2. The monocular distance measuring method based on microlens array digital imaging according to claim 1, wherein: the calculating the sharpness of the different partitions of each sample refocused image comprises:
and calculating a digital imaging focal plane corresponding to the clearest partition in each sample refocusing image by using a digital focusing algorithm.
3. The monocular distance measuring method based on microlens array digital imaging of claim 2, wherein: the calculating the sharpness of the different regions of each sample refocused image further comprises:
and performing definition calculation on all the partitions in each sample refocusing image by using a definition evaluation function to obtain a definition curve of all the partitions in each sample refocusing image.
4. The monocular distance measuring method based on microlens array digital imaging of claim 3, wherein: the method for obtaining the ranging distance of the clearest partition comprises the following steps:
and determining the clearest partition according to the definition curve of each sample refocusing image, and calculating to obtain the distance measurement distance corresponding to the current digital refocusing parameter according to the digital imaging focus plane corresponding to the clearest partition.
5. Monocular distance measuring method based on digital imaging of microlens arrays according to any of the claims 1 to 4, characterized in that: the partition method of different partitions of the sample refocused image comprises the following steps:
the tilt directions corresponding to the sample refocused image are equally divided into a plurality of partitions.
6. Monocular distance measuring method based on digital imaging of microlens arrays according to any of the claims 3 or 4, characterized in that:
the method for obtaining the digital imaging focal plane comprises the following steps:
according to the formula of the digital focusing algorithm, calculating to obtain a digital imaging focal plane image E corresponding to the clearest partition in the sample refocusing image corresponding to the digital refocusing parameter alphaαF(x,y)：
In the formula LαF(u, v, x, y) is the spectral energy carried by the point (x, y) on the x-y plane through which the light ray on the F image plane passes from the point (u, v) on the u-v plane at the image distance l ═ F; the u-v plane represents the plane of the micro lens array, and the x-y plane represents the plane of the detector; l isFRepresenting the path after entering a microlens array cameraThe point (u, v) on the u-v plane is projected to the spectral energy carried by the point (x, y) on the x-y plane.
7. The monocular distance measuring method based on microlens array digital imaging of claim 6, wherein: the method for performing definition calculation on all the partitions in each sample refocused image by using the definition evaluation function comprises the following steps:
the sharpness evaluation result f (i) for each partition was calculated using the following formula:
wherein C is a gradient matrix.
8. The monocular distance measuring method based on microlens array digital imaging of claim 7, wherein: the gradient matrix C is:
C＝I*L，
in the formula, I is the gray value corresponding to the pixels of different partitions in the sample refocusing image, and the digital imaging focal plane image E corresponding to different partitionsαF(x, y); the L is a Laplacian operator;
when the sample refocused image of the scale is divided equally into 60 sections along the scale direction:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910210286.9A CN109883391B (en) | 2019-03-20 | 2019-03-20 | Monocular distance measurement method based on digital imaging of microlens array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910210286.9A CN109883391B (en) | 2019-03-20 | 2019-03-20 | Monocular distance measurement method based on digital imaging of microlens array |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109883391A CN109883391A (en) | 2019-06-14 |
| CN109883391B true CN109883391B (en) | 2021-09-24 |
Family
ID=66933094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910210286.9A Active CN109883391B (en) | 2019-03-20 | 2019-03-20 | Monocular distance measurement method based on digital imaging of microlens array |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109883391B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110455253A (en) * | 2019-09-03 | 2019-11-15 | 北京格如灵科技有限公司 | Direct reading outline range-measurement system and method based on binocular parallax under a kind of virtual reality and augmented reality environment |
| CN111238403A (en) * | 2020-01-18 | 2020-06-05 | 武汉盛信鸿通科技有限公司 | Three-dimensional reconstruction method and device based on light field sub-aperture stripe image |
| CN111696116B (en) * | 2020-04-30 | 2024-07-12 | 江苏橙智云信息技术有限公司 | Method for removing image shielding object by using camera array |
| CN113838075B (en) * | 2020-06-23 | 2024-01-09 | 南宁富联富桂精密工业有限公司 | Monocular ranging method, monocular ranging device and computer readable storage medium |
| CN113124819B (en) * | 2021-06-17 | 2021-09-10 | 中国空气动力研究与发展中心低速空气动力研究所 | Monocular distance measuring method based on plane mirror |
| CN114674276B (en) * | 2022-03-25 | 2024-02-23 | 南京汇川图像视觉技术有限公司 | Distance measurement method, machine vision system, and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1419147A (en) * | 2001-11-13 | 2003-05-21 | 中国科学院力学研究所 | Automatic focusing regulation method and device for multiple variable ptial imaging system of 3D object |
| CN102761700A (en) * | 2011-04-29 | 2012-10-31 | 国际商业机器公司 | Shooting device and method for obtaining distance between different points on shot object |
| CN106373152A (en) * | 2016-09-18 | 2017-02-01 | 清华大学深圳研究生院 | Distance estimation method based on hand-held light field camera |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI491856B (en) * | 2014-06-20 | 2015-07-11 | Univ Nat Central | Detecting system |
-
2019
- 2019-03-20 CN CN201910210286.9A patent/CN109883391B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1419147A (en) * | 2001-11-13 | 2003-05-21 | 中国科学院力学研究所 | Automatic focusing regulation method and device for multiple variable ptial imaging system of 3D object |
| CN102761700A (en) * | 2011-04-29 | 2012-10-31 | 国际商业机器公司 | Shooting device and method for obtaining distance between different points on shot object |
| CN106373152A (en) * | 2016-09-18 | 2017-02-01 | 清华大学深圳研究生院 | Distance estimation method based on hand-held light field camera |
Non-Patent Citations (4)
| Title |
|---|
| Optical Sectioning Tomographic Reconstruction of Three-Dimensional Flame Temperature Distribution Using Single Light Field Camera;Wenchao Zhao;《IEEE SENSORS JOURNAL》;20180115;第18卷(第2期);第528-539页 * |
| 基于光场分层成像的火焰三维温度场测量;赵文超等;《东南大学学报(自然科学版)》;20170930;第47卷(第5期);全文 * |
| 基于光场摄像技术的对焦测距方法的研究;肖相国等;《光子学报》;20081231;第37卷(第12期);第2539-2543页 * |
| 数字图像清晰度评价函数的研究与改进;蒋金银;《电子世界》;20130923(第16期);第224-226页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109883391A (en) | 2019-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109883391B (en) | Monocular distance measurement method based on digital imaging of microlens array | |
| EP3516626B1 (en) | Device and method for obtaining distance information from views | |
| JP6456156B2 (en) | Normal line information generating apparatus, imaging apparatus, normal line information generating method, and normal line information generating program | |
| CN107084794B (en) | Flame three-dimensional temperature field measuring system and method based on light field layered imaging technology | |
| US8305485B2 (en) | Digital camera with coded aperture rangefinder | |
| US20070189750A1 (en) | Method of and apparatus for simultaneously capturing and generating multiple blurred images | |
| EP1580523A1 (en) | Three-dimensional shape measuring method and its device | |
| CN105282443B (en) | A kind of panorama depth panoramic picture imaging method | |
| JP6786225B2 (en) | Image processing equipment, imaging equipment and image processing programs | |
| JP7378219B2 (en) | Imaging device, image processing device, control method, and program | |
| JP2023115356A (en) | Measurement device, imaging device, control method, and program | |
| JPH1069543A (en) | Method and device for reconstituting curved surface of object | |
| CN113808019A (en) | Non-contact measurement system and method | |
| JP2017134561A (en) | Image processing device, imaging apparatus and image processing program | |
| RU2734070C1 (en) | Method of measuring spatial distance between small objects | |
| JPH0252204A (en) | Measuring instrument for three-dimensional coordinate | |
| JP7373297B2 (en) | Image processing device, image processing method and program | |
| Shafer | Automation and calibration for robot vision systems | |
| RU2806249C1 (en) | Method for obtaining high spatial resolution images by opto-electronic observation tool for remote objects | |
| Fuse et al. | 3D measurement combining multi-view and multi-focus images using light field camera | |
| WO2019012660A1 (en) | Image processing device and light field imaging device | |
| Zhang et al. | Image Acquisition Devices | |
| CN107726993B (en) | Particle depth measurement method based on light field image macro pixel greatest gradient region | |
| Marrocchesi et al. | A Lens Characterization Method for Low-budget High-quality Museum Photography | |
| Ho et al. | Design and Implementation of a Hand-held Lensless Light Field Camera |
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 |