CN107680083B - Parallax determination method and parallax determination device - Google Patents
Parallax determination method and parallax determination device Download PDFInfo
- Publication number
- CN107680083B CN107680083B CN201710823614.3A CN201710823614A CN107680083B CN 107680083 B CN107680083 B CN 107680083B CN 201710823614 A CN201710823614 A CN 201710823614A CN 107680083 B CN107680083 B CN 107680083B
- Authority
- CN
- China
- Prior art keywords
- parallax
- image
- matching
- edge
- pixel points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/97—Determining parameters from multiple pictures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
Abstract
The invention relates to a parallax determining method and a device, wherein the method comprises the following steps: determining a reference edge binary image and a matching edge binary image of a scene; determining a coarse matching parallax image according to the pixel points with the effective parallax and the corresponding effective parallax; determining the parallax ranges of the non-edge pixel points and the edge pixel points according to the effective parallax of the non-edge pixel points and the edge pixel points in the reference image in the coarse matching parallax image; calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the gray-scale values of the pixel points in the matched image; and determining a corresponding target parallax when the value of the cost function is minimum, and generating a fine matching parallax image according to the reference image and the target parallax. According to the embodiment of the invention, the parallax range of the matching window moving in the matching image aiming at partial pixel points is reduced, and the parallax of different pixel points in the whole reference image can be determined more quickly.
Description
Technical Field
The present invention relates to the field of stereo matching technologies, and in particular, to a disparity determination method and a disparity determination apparatus.
Background
In the stereo matching process, in order to determine the parallax of the matched image relative to the reference image, a reference window can be set for each pixel point in the reference image, a matching window can be set for each pixel point in the matched image, a cost function is determined according to the gray scale values of the pixel points in the two windows, then the matching window is moved within a certain parallax range in the matched image, the parallax corresponding to the maximum value of the determined cost function is used as the parallax of the pixel point in the reference window, and then the parallax of each pixel point in the reference image can be determined, so that the parallax image is generated.
However, at present, for each pixel point in the reference image, the parallax ranges selected by moving the matching window in the matching image are all the same, but actually, the feature significance of each pixel point in the reference image is different, and in the related art, different pixel points are not treated differently to select a suitable parallax range to move the matching window, which results in lower efficiency of determining the parallax.
Disclosure of Invention
The invention provides a parallax determination method and a parallax determination device, which are used for solving the defects in the related art.
According to a first aspect of embodiments of the present invention, there is provided a disparity determining method, including:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
and determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image.
According to a second aspect of embodiments of the present invention, there is provided a parallax determining apparatus comprising:
the binary image determining unit is used for determining a reference edge binary image and a matching edge binary image of the same scene based on the reference image and the matching image of the same scene shot by the two cameras;
the rough matching unit is used for determining a rough matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
the range determining unit is used for determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
the cost calculation unit is used for calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the gray-scale values of the pixel points in the matched image;
and the fine matching unit is used for determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image.
According to a third aspect of the embodiments of the present invention, there is provided a terminal, including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
and determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image.
According to a fourth aspect of embodiments of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
and determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image.
According to the embodiment, the first parallax range of the matching window is moved in the matching window for the non-edge pixel points, the second parallax range of the matching window is moved in the matching window for the matched pixel points with effective parallax, and the first parallax range and the second parallax range are both smaller than the preset parallax range, so that the parallax range of the matching window is moved in the matching image for part of the pixel points, the parallax of different pixel points in the whole reference image can be determined more quickly, the parallax determining efficiency is improved, the parallax image generating efficiency is improved, and the matching accuracy is improved to a certain extent.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 shows a schematic flow diagram of a disparity determination method according to an embodiment of the present invention.
Fig. 2 is a schematic diagram illustrating a filtering template according to an embodiment of the present invention.
Fig. 3 is a schematic diagram illustrating a source image according to an embodiment of the present invention.
Fig. 4 is a schematic diagram illustrating an edge binary image according to an embodiment of the present invention.
Fig. 5 is a schematic diagram illustrating a window in accordance with one embodiment of the present invention.
Fig. 6 shows a schematic flow diagram of determining a coarse matching parallax image of a reference image according to one embodiment of the present invention.
Fig. 7 shows a schematic flow diagram of determining a disparity range according to an embodiment of the present invention.
Fig. 8 shows another schematic flow diagram for determining a disparity range according to an embodiment of the invention.
Fig. 9 shows a schematic flow diagram of another disparity determination method according to an embodiment of the present invention.
Fig. 10 shows a schematic flow diagram of yet another disparity determination method according to an embodiment of the present invention.
FIG. 11 is a diagram illustrating a reference image, according to one embodiment of the present invention.
FIG. 12 is a diagram illustrating matching images, according to one embodiment of the present invention.
Fig. 13 is a diagram illustrating edge-based stereo matching disparity according to an embodiment of the present invention.
Fig. 14 is a schematic block diagram illustrating a parallax determination apparatus according to an embodiment of the present invention.
Fig. 15 is a schematic block diagram showing another parallax determining apparatus according to an embodiment of the present invention.
Fig. 16 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention.
Fig. 17 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention.
Fig. 18 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention.
Fig. 19 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the invention, as detailed in the appended claims.
Fig. 1 shows a schematic flow diagram of a disparity determination method according to an embodiment of the present invention. Fig. 1 therefore, the method comprises:
in step S11, a reference edge binary image and a matching edge binary image of the scene are determined based on the reference image and the matching image of the same scene captured by the two cameras.
In one embodiment, the images may be acquired by two image acquisition devices, such as cameras, respectively, with the image acquired by one image acquisition device being the reference image and the image acquired by the other image acquisition device being the matching image.
In an embodiment, for the reference image and the matching image, marginalization processing may be performed first to determine edge pixel points and non-edge pixel points in the image, and then binarization processing is performed on the two types of pixel points, so as to obtain a reference edge binary image according to the reference image, and obtain a matching edge binary image according to the matching image.
In one embodiment, before performing the marginalization and binarization processing, filtering and smoothing processing may be further performed on the reference image and the matching image to remove noise pixel points in the image. The reference view and the matching view may be respectively subjected to filtering smoothing by using gaussian filtering.
Fig. 2 is a schematic diagram illustrating a filtering template according to an embodiment of the present invention.
In an embodiment, a 3 × 3 filtering template as shown in fig. 2 may be used to perform gaussian filtering on each pixel point in the reference image and the comparison image, and for convenience, the filtered reference image is not denoted as imgR, and the filtered comparison image is denoted as imgL.
In one embodiment, the edge binarization processing may specifically include two steps:
firstly, a reference edge pixel point is extracted from a reference image, and a matched edge pixel point is extracted from a matched image, wherein a canny operator can be used in the process of extracting the edge pixel point, and a Sobel operator, an Isotropic Sobel operator, a Roberts operator, a Prewitt operator and a Laplacian operator can also be used in the process of extracting the edge pixel point.
And secondly, carrying out binarization processing on the reference edge pixel points and carrying out binarization processing on the matched edge pixel points, wherein the binarization processing process can use a canny operator and also can use a Kirsch operator.
Preferably, the above two operations can be done according to the canny operator.
Fig. 3 is a schematic diagram illustrating a source image according to an embodiment of the present invention.
As shown in fig. 3, the left image is a left image captured by one image capturing device, and the right image is a right image captured by another image capturing device, wherein the left image can be used as a reference image, and the right image can be used as a matching image.
Fig. 4 is a schematic diagram illustrating an edge binary image according to an embodiment of the present invention.
As shown in fig. 4, the left image is a reference edge binary image obtained by performing edge binarization processing on the reference image in fig. 3, and the right image is a matched edge binary image obtained by performing edge binarization processing on the reference image in fig. 3.
In step S12, a coarse matching disparity image of the reference image is determined according to the pixel points of the reference edge binary image and the matching edge binary image having effective disparity and the corresponding effective disparity.
In one embodiment, a reference binary pixel point can be determined in a reference edge binary image, a reference binary window is determined by taking the reference binary pixel point as a center, a matching binary pixel point is determined in a matching binary image, and a matching binary window is determined by taking the matching binary pixel point as a center.
In one embodiment, the reference window and the window to be matched may be n × n windows, that is, windows including n × n pixel points.
In one embodiment, the pixel points in the reference edge binary image are reference binary pixel points, the reference binary pixel points include reference edge binary pixel points and reference non-edge binary pixel points, and correspondingly, the matching binary pixel points include matching edge binary pixel points and matching non-edge binary pixel points. And determining a reference binary window only for the reference edge binary pixel points, and not processing the reference non-edge binary pixel points.
Fig. 5 is a schematic diagram illustrating a window in accordance with one embodiment of the present invention.
In one embodiment, as shown in fig. 5, the left image is a reference edge binary image, where a gray rectangular frame is a reference window, and the reference binary pixel point is located in the center of the reference window. The right image is a matching edge binary image, a gray rectangular frame is a window to be matched, a matching binary pixel point is located in the center of the matching binary window, and a white rectangular frame indicates that the matching binary window is moved within a preset parallax range. Of course, in the actual application process, the size of the window and the preset parallax range may be set as needed, the size of the window may be different from that shown in fig. 5, and the window movement range may also be different from that shown in fig. 5.
In one embodiment, the matching binary window may be moved within a preset parallax range in the matching edge binary image, and the number N of pixel points in the matching binary window, which are 1 from the value of the corresponding pixel point in the reference binary window, is determinedrAnd the number N of pixel points having different values corresponding to the pixel pointswCalculating
Determining the maximum value C of CmaxWhether the effective parallax is larger than the preset cost value or not is judged, if so, the effective parallax of the pixel points corresponding to the reference binary window is recorded, and the effective parallax is CmaxAnd if the corresponding parallax is not larger than the corresponding parallax, recording that the pixel point corresponding to the reference binary window does not have effective parallax. And then determining a coarse matching parallax image according to the pixel points with the effective parallax in the reference edge binary image and the corresponding effective parallax.
Wherein, the larger C is, it indicates that the positions (which are positions in the window but not necessarily positions in the image) in the matching binary window and the reference binary window are the same, and are both edge binary pixel points (that is, a pixel point at a certain position in the matching binary window is a matching edge binary pixel point, and a pixel at the same position in the reference binary window is a pixel at the same position in the matching binary windowThe point is a reference edge binary pixel point) has a larger proportion in the window, that is, the matching degree between the matching binary window and the reference binary window is higher, so that a preset cost value can be set, and if the maximum value C of C is determined in a preset parallax range, the maximum value C of C is determinedmaxSufficiently large, i.e. greater than a predetermined cost value, then C can be determined to a greater extentmaxThe corresponding parallax is the parallax of the pixel point corresponding to the reference binary window, so that the effective parallax of the pixel point can be recorded. Accordingly, if CmaxIs not large enough, it is difficult to ensure CmaxAnd if the corresponding parallax is the parallax of the pixel point corresponding to the reference binary window, recording that the pixel point has no effective parallax.
It should be noted that the denominator C may include the number of pixel points in the matching binary window, which are equal to 1 in the value of the corresponding pixel point in the reference binary window, and the number of pixel points with different values of the corresponding pixel point, and may also include the number of pixel points with 1 in the value of the corresponding pixel point.
In step S13, determining the parallax range of the non-edge pixel point and the edge pixel point of the reference image according to the effective parallax of the non-edge pixel point and the edge pixel point in the coarse matching parallax image.
In one embodiment, the determination of the parallax range is different for non-edge pixels and edge pixels.
For the non-edge pixel points, the non-edge pixel points can be determined in the reference image according to the reference non-edge binary pixel points in the reference edge binary image, then the left effective parallax of the left edge pixel point which is nearest to the left side of the non-edge pixel point and has the effective parallax and the right effective parallax of the right edge pixel point which is nearest to the right side and has the effective parallax are determined in the rough matching parallax image, and a first parallax range is determined according to the left effective parallax and the right effective parallax, wherein the first parallax range is within the preset parallax range.
For the edge pixel points, the edge pixel points can be determined in the reference image according to the reference edge binary pixel points in the reference edge binary image, and the pixel points with effective parallax and the effective parallax thereof in the coarse matching parallax image are determined in the edge pixel points.
Further, for an edge pixel point with an effective parallax, a second parallax range may be determined according to the effective parallax and a second preset parallax threshold, where the second parallax range is within the preset parallax range. Besides the edge pixel points with the effective parallax, the edge pixel points without the effective parallax can be determined, and the parallax range of the edge pixel points without the effective parallax can be determined to be the preset parallax range.
In step S14, a value of a cost function of a pixel point in the reference image is calculated according to the parallax range, the gray scale values of the pixel points in the reference image and the matching image.
In one embodiment, for a non-edge pixel point, if a left effective parallax and a right effective parallax can be determined in the rough matching parallax image, the matching window is moved within a first parallax range in the matching image, otherwise, the matching window is moved within the preset parallax range in the matching image.
And moving the matching window in a second parallax range in the matching image aiming at the edge pixel points with effective parallax.
And moving a matching window within the preset parallax range in the matching image aiming at the unmatched edge pixel points.
Based on the method, for the non-edge pixel points, the edge pixel points with effective parallax and the unmatched edge pixel points, different parallax ranges can be respectively determined in the matched images to move the matching windows, so that the pixel points matched with the pixel points in the reference image in the matched images are accurately determined.
In one embodiment, the cost function may be a similarity cost function, such as a zero mean gray and correlation coefficient function (ZCC), a normalized gray cross-correlation function (NCC), and the following formulas respectively:
ZCC and NCC can be used to calculate gray level similarity.
The cost function may also be a difference cost function, such as a Sum of Absolute Differences (SAD) of pixel points, a Sum of Square Difference (SSD) of pixel points, etc., and the formula is as follows:
SAD and SSD may be used to calculate gray scale differences.
In the above formula, W is the window, I1(x, y) represents the gray scale of a pixel point with coordinates (x, y) in the reference binary image, I2(x + d, y) represents the gray scale of the pixel point with the coordinate (x + d, y) in the matching binary image, d represents the parallax, avgW1Representing the mean value of the gray levels of the pixels in the reference window, avgW2And representing the mean value of the pixel points in the window to be matched.
Under the condition that the cost function is a similarity cost function, the corresponding parallax when the value of the cost function is maximum is the parallax value of the pixel point; and under the condition that the cost function is the difference cost function, the corresponding parallax when the value of the cost function is minimum is the parallax value of the pixel point.
In step S15, the disparity value with the minimum value of the cost function is determined as the disparity value of the pixel point in the reference image.
In an embodiment, a disparity value may be determined for each pixel point in the reference image, and then the display effect of the pixel point is adjusted according to the disparity value corresponding to each pixel point, for example, the pixel point with the larger disparity value is displayed as red, and the pixel point with the smaller disparity value is displayed as blue, so that the disparity image can be obtained.
In one embodiment, for a non-edge pixel point, a first parallax range of a matching window is moved in the matching window, and for an edge pixel point with effective parallax, a second parallax range of the matching window is moved in the matching window and is smaller than a preset parallax range, so that the parallax range of the matching window moved in the matching image for a part of pixel points is reduced, the parallax of different pixel points in the whole reference image can be determined more quickly, the parallax determining efficiency is improved, the parallax image generating efficiency is improved, and the matching accuracy can be improved to a certain extent.
Fig. 6 shows a schematic flow diagram of determining a coarse matching parallax image of a reference image according to one embodiment of the present invention. As shown in fig. 6, the determining a coarse matching parallax image of the reference image according to the pixel points having effective parallax and the corresponding effective parallax in the reference edge binary image and the matching edge binary image includes:
in step S121, the number N of pixels whose pixel values of the pixel points of the matching edge binary image in the matching binary window and the pixel points of the reference edge binary image in the reference binary window are both 1 is determinedr；
In step S122, the number N of pixels having different pixel values between the pixel point of the matching edge binary image in the matching binary window and the corresponding pixel point of the reference edge binary image in the reference binary window is determinedw；
In step S123, the matching binary window is moved within the preset parallax range in the matching edge binary image according to the formula
Determining the maximum value C of Cmax；
In step S124, if CmaxIf the value is greater than the preset cost value, determining that the parallax value of the pixel point corresponding to the reference binary window is effective parallax, and the effective parallax is CmaxA corresponding parallax;
in step S125, ifCmaxIf the current value is not greater than the preset cost value, determining the parallax value of the pixel point corresponding to the reference binary window as an invalid parallax;
in step S126, a coarse matching parallax image of the reference image is determined according to the pixel points having effective parallax and the corresponding effective parallax in the reference edge binary image and the matching edge binary image.
In one embodiment, the larger C is, it indicates that the positions in the matching binary window and the reference binary window are the same, and the ratio of the pixel points that are edge binary pixel points is larger in the window, that is, it indicates that the matching degree of the matching binary window and the reference binary window is higher.
Therefore, a predetermined cost value can be set, if the maximum value C of C is determined within a predetermined parallax rangemaxIs sufficiently large, i.e. CmaxGreater than a predetermined cost value, then C can be determined to a greater extentmaxThe corresponding parallax is the parallax of the pixel point corresponding to the reference binary window, so that the effective parallax of the pixel point can be recorded. Accordingly, if CmaxIs not large enough, i.e. CmaxIf it is not greater than the predetermined cost value, it is difficult to ensure CmaxAnd if the corresponding parallax is the parallax of the pixel point corresponding to the reference binary window, recording that the pixel point has no effective parallax.
Further, for the pixel points with effective parallax, the information of the effective parallax can be attached to the corresponding pixel points, for example, the pixel points are set to have different colors according to different effective parallaxes, so that a parallax image formed by the pixel points with the effective parallax of the reference edge binary image can be obtained, and the parallax image is also roughly matched.
Fig. 7 shows a schematic flow diagram of determining a disparity range according to an embodiment of the present invention. As shown in fig. 7, the determining, according to the effective parallax of the non-edge pixel point and the edge pixel point in the coarse matching parallax image, the parallax range of the non-edge pixel point and the edge pixel point of the reference image includes:
in step S131, if a left effective parallax and a right effective parallax exist in an adjacent pixel of a non-edge pixel in the reference image in the rough matching parallax image, determining a first parallax range of the non-edge pixel according to the left effective parallax and the right effective parallax, where the first parallax range is within a preset parallax range;
in step S132, if there is no left effective parallax and no right effective parallax between non-edge pixel points in the reference image and adjacent pixel points in the rough matching parallax image, determining that the parallax range of the non-edge pixel points is the preset parallax range.
In one embodiment, for a non-edge pixel point in the reference image, a left effective parallax of a left edge pixel point nearest to the left side of the non-edge pixel point and having an effective parallax, and a right effective parallax of a right edge pixel point nearest to the right side and having an effective parallax are determined in the rough matching parallax image;
for a non-edge pixel point in an object in the reference image, the pixel point corresponding to the non-edge pixel point in the matching image is also positioned in the object, namely positioned in the edge of the object, but not positioned outside the edge of the object. Based on this, for the non-edge pixel point in the reference image, the nearest edge pixel point on the left side (i.e., the left edge pixel point) and the nearest edge pixel point on the right side (i.e., the right edge pixel point) can be determined, and if the left edge pixel point and the right edge pixel point exist, the effective parallax (i.e., the left effective parallax) corresponding to the reference edge binary pixel point corresponding to the left edge pixel point in the reference edge binary image can be further determined, and the effective parallax (i.e., the right effective parallax) corresponding to the reference edge binary pixel point corresponding to the right edge pixel point in the reference edge binary image can be further determined. Then, in the matching image, the matched pixel point determined for the pixel point should also be between the disparity ranges defined by the corresponding left and right effective disparities in the matching image.
Correspondingly, if there is no left effective parallax and no right effective parallax between non-edge pixel points in the reference image and adjacent pixel points in the rough matching parallax image, that is, there is no left edge pixel point and no right edge pixel point, or even if there is one side edge pixel point, the parallax of the side edge pixel point is an invalid parallax, that is, the parallax range of the matched pixel point in the matching image cannot be determined according to the left effective parallax and the right effective parallax, and then the parallax range of the non-edge pixel point can be determined to be the preset parallax range.
And because the left edge pixel point and the right edge pixel point are both located in the matching image, the determined first parallax range is smaller than the preset parallax range corresponding to the moving matching window in the whole matching image according to the boundary of the first parallax range, and the matching speed can be improved by moving the matching window according to the first parallax range.
The steps S131 and S132 may be performed in the order shown in fig. 7, or may be performed in another order, for example, in parallel, with the execution order being adjusted as necessary.
Optionally, the determining a first disparity range of the non-edge pixel point according to the left effective disparity and the right effective disparity includes:
according to the left effective parallax dLAnd the right effective parallax dRAnd a first predetermined parallax threshold dT1A first range of disparity is determined and,
at dL<dRIn the case of (1), the first parallax range is [ d ]L-dT1，dR+dT1]Wherein the predetermined parallax range is [ d ]min，dmax]，dL-dT1Is greater than or equal to dmin，dR+dT1Is less than or equal to dmax；
At dL>dRIn the case of (1), the first parallax range is [ d ]R-dT1，dL+dT1]Wherein the predetermined parallax range is [ d ]min，dmax]，dR-dT1Is greater than or equal to dmin，dL+dT1Is less than or equal to dmax。
Fig. 8 shows another schematic flow diagram for determining a disparity range according to an embodiment of the invention. As shown in fig. 8, the determining, according to the effective parallax of the non-edge pixel point and the edge pixel point in the reference image in the coarse matching parallax image, the parallax range of the non-edge pixel point and the edge pixel point includes:
in step S133, if there is a matching edge pixel in the coarse matching parallax image among edge pixels in the reference image, determining a second parallax range of the matching edge pixel according to the effective parallax of the matching edge pixel and a second preset parallax threshold, where the second parallax range is within a preset parallax range;
in step S134, if there is no matching edge pixel in the coarse matching parallax image for the edge pixel in the reference image, it is determined that the second parallax range of the matching edge pixel is the preset parallax range.
In one embodiment, for an edge pixel, according to the embodiment shown in fig. 6, it may be recorded whether there is an effective disparity, and a specific effective disparity in the case that there is an effective disparity. Aiming at the edge pixel points with effective parallax, the effective parallax of the pixel points in the reference image is also determined roughly, wherein the effective parallax of the pixel points is either the effective parallax or the parallax corresponding to the matched pixel points within a smaller range from the effective parallax. The second disparity range can thus be determined from the effective disparity and the second preset disparity threshold.
The steps S133 and S134 may be executed in the order shown in fig. 8, or may be executed in another order, for example, in parallel.
Optionally, the determining, according to the effective disparity and a second preset disparity threshold, a second disparity range of the matching edge pixel point includes:
according to the effective parallax d and a second preset parallax threshold dT2Determining a second disparity range d-dT2，d+dT2]Wherein the predetermined parallax range is [ d ]min，dmax]，d-dT2Is greater than or equal to dmin，d+dT2Is less than or equal to dmax。
In an embodiment, for an edge pixel, if it is recorded according to the embodiment shown in fig. 6 that there is no effective parallax, it is described that the method for determining the matching degree according to the embodiment shown in fig. 6 does not determine a pixel that matches a pixel in the reference image in the matching binary image, and then when performing fine matching in the matching image, it is necessary to re-determine the pixel that matches the edge pixel within the preset parallax range corresponding to the entire matching image, so as to determine the parallax corresponding to the edge pixel.
Fig. 9 shows a schematic flow diagram of another disparity determination method according to an embodiment of the present invention. As shown in fig. 9, the calculating the value of the cost function of the pixel point in the reference image includes:
in step S141, the sum C of cost function values of the non-edge pixels is calculated according to the gray scale values of the pixels in the reference window and the gray scale values of the pixels in the matching windowS(x,y)And the sum of the cost function values C of the edge pixelsS’(x,y)；
In step S142, a cost function C is calculated(x,y)＝λCS’(x,y)+(1-λ)CS(x,y)A value of (b), wherein, 0.5<λ<1。
In one embodiment, since 0.5< λ <1, (1- λ) < λ, that is, the weight of the edge pixel in the matching window is larger, and the weight of the non-edge pixel is smaller. In general, when a cost function is calculated, the cost function value of a pixel point with relatively obvious characteristics (for example, an edge pixel point) is mainly concerned, so that a higher weight is set for the edge pixel point, the weight of the pixel point with more obvious characteristics (also called the edge pixel point) in the image in the cost function is improved, and the accuracy of calculating the cost function is further improved.
Fig. 10 shows a schematic flow diagram of yet another disparity determination method according to an embodiment of the present invention. As shown in fig. 10, after generating the parallax image, the method further includes:
in step S16, performing filtering smoothing processing on edge pixel points in the parallax image;
in step S17, an edge-based stereo matching parallax image is generated according to the target parallax of the edge pixel in the reference binary image and the edge pixel after the filtering smoothing process.
In an embodiment, only the edge pixel points in the parallax image may be filtered, and only the edge pixel points after filtering and smoothing are extracted to determine the target parallax, so that a parallax image only having the edge pixel points, that is, a stereo matching parallax image based on edges, may be obtained.
FIG. 11 is a diagram illustrating a reference image, according to one embodiment of the present invention. FIG. 12 is a diagram illustrating matching images, according to one embodiment of the present invention. Fig. 13 is a diagram illustrating edge-based stereo matching disparity according to an embodiment of the present invention.
In one embodiment, according to the above embodiments, the reference image (e.g., left image) shown in fig. 11 and the matching image (e.g., right image) shown in fig. 12 may be processed, so as to accurately determine edge pixel points belonging to an edge, and then accurately generate the edge-based stereo matching disparity map shown in fig. 13 according to the edge pixel points.
The invention also provides an embodiment of a parallax determination device corresponding to the embodiment of the parallax determination method.
Fig. 14 is a schematic block diagram illustrating a parallax determination apparatus according to an embodiment of the present invention. As shown in fig. 14, the parallax determining apparatus includes:
a binary image determining unit 1, configured to determine a reference edge binary image and a matching edge binary image of a same scene based on a reference image and a matching image of the scene captured by two cameras;
the rough matching unit 2 is used for determining a rough matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
the range determining unit 3 is configured to determine a parallax range of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
the cost calculation unit 4 is configured to calculate a value of a cost function of a pixel point in the reference image according to the parallax range, the gray scale values of the pixel points in the reference image and the matching image;
and the fine matching unit 5 is configured to determine the disparity value with the minimum value of the cost function as the disparity value of the pixel point in the reference image.
Fig. 15 is a schematic block diagram showing another parallax determining apparatus according to an embodiment of the present invention. As shown in fig. 15, the coarse matching unit 2 includes:
a number determining subunit 21, configured to determine the number N of pixels whose pixel values of the pixel points of the matching edge binary image in the matching binary window and the pixel points of the reference edge binary image in the reference binary window are both 1r(ii) a And determining the number N of pixel points with different pixel values of the pixel points of the matching edge binary image in the matching binary window and the corresponding pixel points of the reference edge binary image in the reference binary windoww；
A cost value subunit 22, configured to move the matching binary window within a preset parallax range in the matching edge binary image according to a formula
Determining the maximum value C of Cmax；
And the parallax sub-unit 23 is configured to determine a coarse matching parallax image of the reference image according to the pixel points having effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image.
Fig. 16 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention. As shown in fig. 16, the range determining unit 3 includes:
a first range determining subunit 31, configured to determine, if a left effective parallax and a right effective parallax exist in an adjacent pixel of a non-edge pixel in the reference image in the coarse matching parallax image, a first parallax range of the non-edge pixel according to the left effective parallax and the right effective parallax, where the first parallax range is within a preset parallax range; and if the adjacent pixel points of the non-edge pixel points in the reference image in the rough matching parallax image do not have the left effective parallax and the right effective parallax, determining the parallax range of the non-edge pixel points as the preset parallax range.
Optionally, the range determining subunit is configured to determine the left effective disparity d according to the left effective disparity dLAnd the right effective parallax dRAnd a first predetermined parallax threshold dT1A first range of disparity is determined and,
at dL<dRIn the case of (1), the first parallax range is [ d ]L-dT1，dR+dT1]Wherein the predetermined parallax range is [ d ]min，dmax]，dL-dT1Is greater than or equal to dmin，dR+dT1Is less than or equal to dmax；
At dL>dRIn the case of (1), the first parallax range is [ d ]R-dT1，dL+dT1]Wherein the predetermined parallax range is [ d ]min，dmax]，dR-dT1Is greater than or equal to dmin，dL+dT1Is less than or equal to dmax。
Fig. 17 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention. As shown in fig. 17, the range determination unit 3 includes:
a second range determining subunit 32, configured to determine, if there is a matching edge pixel in the coarse matching parallax image among edge pixels in the reference image, a second parallax range of the matching edge pixel according to an effective parallax of the matching edge pixel and a second preset parallax threshold, where the second parallax range is within a preset parallax range; and if the edge pixel points in the reference image do not have matched edge pixel points in the coarse matching parallax image, determining that a second parallax range of the matched edge pixel points is a preset parallax range.
Optionally, the second range determination subunit is configured to determine the second range according to the effective disparity d and a second preset disparity threshold dT2Determining a second disparity range d-dT2，d+dT2]Wherein the predetermined parallax range is [ d ]min，dmax]，d-dT2Is greater than or equal to dmin，d+dT2Is less than or equal to dmax。
Fig. 18 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention. As shown in fig. 18, the cost calculation unit 4 includes:
a cost sum subunit 41, configured to calculate a sum C of cost function values of the non-edge pixel points according to the gray-scale values of the pixels in the reference window and the gray-scale values of the pixels in the matching windowS(x,y)And the sum of the cost function values C of the edge pixelsS’(x,y)；
A weighting subunit 42 for calculating a cost function C(x,y)＝λCS’(x,y)+(1-λ)CS(x,y)A value of (b), wherein, 0.5<λ<1。
Fig. 19 is a schematic block diagram illustrating still another parallax determining apparatus according to an embodiment of the present invention. As shown in fig. 19, the method further includes:
the smoothing unit 7 is configured to perform filtering smoothing on edge pixel points in the parallax image;
and the image generating unit 8 is configured to generate an edge-based stereo matching parallax image according to the target parallax of the edge pixel in the reference binary image and the edge pixel after the filtering smoothing processing.
With regard to the apparatus in the above-described embodiments, the specific manner in which each unit performs operations has been described in detail in the embodiments of the related method, and will not be described in detail here.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of the invention. One of ordinary skill in the art can understand and implement it without inventive effort.
An embodiment of the present invention relates to a terminal, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
and determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image.
An embodiment of the invention relates to a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
and determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.
Claims (8)
1. A disparity determination method, comprising:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image, including:
determining the number N of pixel points of which the pixel values of the pixel points of the matching edge binary image in the matching binary window and the pixel points of the reference edge binary image in the reference binary window are both 1r；
Determining the number N of pixel points with different pixel values of the pixel points of the matching edge binary image in the matching binary window and the corresponding pixel points of the reference edge binary image in the reference binary windoww；
Moving the matching binary window within a preset parallax range in the matching edge binary image according to a formula
Determining the maximum value C of Cmax；
If C is presentmaxIf the value is greater than the preset cost value, determining that the parallax value of the pixel point corresponding to the reference binary window is effective parallax, and the effective parallax is CmaxA corresponding parallax;
if C is presentmaxIf it is not greater than the predetermined cost value, thenDetermining the parallax value of the pixel point corresponding to the reference binary window as an invalid parallax;
and determining a coarse matching parallax image of the reference image according to the pixel points with effective parallax and the corresponding effective parallax in the reference edge binary image and the matching edge binary image.
2. The method of claim 1, wherein determining the disparity ranges of the non-edge pixels and the edge pixels of the reference image according to the effective disparity of the non-edge pixels and the edge pixels in the coarse matching disparity image comprises:
if the adjacent pixel points of the non-edge pixel points in the reference image in the rough matching parallax image have the left effective parallax and the right effective parallax, determining a first parallax range of the non-edge pixel points according to the left effective parallax and the right effective parallax, wherein the first parallax range is within a preset parallax range;
and if the adjacent pixel points of the non-edge pixel points in the reference image in the rough matching parallax image do not have the left effective parallax and the right effective parallax, determining the parallax range of the non-edge pixel points as the preset parallax range.
3. The method of claim 1, wherein determining the disparity ranges of the non-edge pixels and the edge pixels of the reference image according to the effective disparity of the non-edge pixels and the edge pixels in the coarse matching disparity image comprises:
if the edge pixel points in the reference image have matched edge pixel points in the coarse matching parallax image, determining a second parallax range of the matched edge pixel points according to the effective parallax of the matched edge pixel points and a second preset parallax threshold value, wherein the second parallax range is within a preset parallax range;
and if the edge pixel points in the reference image do not have matched edge pixel points in the coarse matching parallax image, determining that a second parallax range of the matched edge pixel points is a preset parallax range.
4. The method of claim 1, wherein calculating the value of the cost function for the pixel in the reference image according to the disparity range, the gray scale values for the pixel in the reference image, and the gray scale values for the pixel in the matching image comprises:
calculating the sum C of cost function values of the non-edge pixel points according to the gray-scale values of the pixel points in the reference window and the gray-scale values of the pixel points in the matching windowS(x,y)And the sum of the cost function values C of the edge pixelsS’(x,y)；
Calculating a cost function C(x,y)＝λCS’(x,y)+(1-λ)CS(x,y)A value of (b), wherein, 0.5<λ<1。
5. A parallax determination apparatus, characterized by comprising:
the binary image determining unit is used for determining a reference edge binary image and a matching edge binary image of the same scene based on the reference image and the matching image of the same scene shot by the two cameras;
the rough matching unit is used for determining a rough matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
the range determining unit is used for determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
the cost calculation unit is used for calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the gray-scale values of the pixel points in the matched image;
the fine matching unit is used for determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image;
wherein the coarse matching unit includes:
a number determining subunit, configured to determine the number N of pixels whose pixel values of the pixel points of the matching edge binary image in the matching binary window and the pixel points of the reference edge binary image in the reference binary window are both 1r(ii) a And determining the number N of pixel points with different pixel values of the pixel points of the matching edge binary image in the matching binary window and the corresponding pixel points of the reference edge binary image in the reference binary windoww；
A cost value subunit for moving the matching binary window within a preset parallax range in the matching edge binary image according to a formula
Determining the maximum value C of Cmax；
And the parallax sub-unit is used for determining a coarse matching parallax image of the reference image according to the pixel points with effective parallax and the corresponding effective parallax in the reference edge binary image and the matching edge binary image.
6. The apparatus of claim 5, wherein the range determination unit comprises:
a first range determining subunit, configured to determine, if a left effective parallax and a right effective parallax exist in an adjacent pixel of a non-edge pixel in the reference image in the coarse matching parallax image, a first parallax range of the non-edge pixel according to the left effective parallax and the right effective parallax, where the first parallax range is within a preset parallax range; and if the adjacent pixel points of the non-edge pixel points in the reference image in the rough matching parallax image do not have the left effective parallax and the right effective parallax, determining the parallax range of the non-edge pixel points as the preset parallax range.
7. A terminal comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program performs the steps of:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image, including:
determining the number N of pixel points of which the pixel values of the pixel points of the matching edge binary image in the matching binary window and the pixel points of the reference edge binary image in the reference binary window are both 1r；
Determining the number N of pixel points with different pixel values of the pixel points of the matching edge binary image in the matching binary window and the corresponding pixel points of the reference edge binary image in the reference binary windoww；
Moving the matching binary window within a preset parallax range in the matching edge binary image according to a formula
Determining the maximum value C of Cmax；
If C is presentmaxIf the value is greater than the preset cost value, determining that the parallax value of the pixel point corresponding to the reference binary window is effective parallax, and the effective parallax is CmaxA corresponding parallax;
if C is presentmaxIf the current value is not greater than the preset cost value, determining the parallax value of the pixel point corresponding to the reference binary window as an invalid parallax;
and determining a coarse matching parallax image of the reference image according to the pixel points with effective parallax and the corresponding effective parallax in the reference edge binary image and the matching edge binary image.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of:
the method comprises the steps of shooting a reference image and a matching image of the same scene based on two cameras, and determining a reference edge binary image and a matching edge binary image of the scene;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image;
determining the parallax ranges of the non-edge pixel points and the edge pixel points of the reference image according to the effective parallax of the non-edge pixel points and the edge pixel points in the coarse matching parallax image;
calculating the value of a cost function of the pixel points in the reference image according to the parallax range, the gray-scale values of the pixel points in the reference image and the matching image;
determining the parallax value with the minimum value of the cost function as the parallax value of the pixel point in the reference image;
determining a coarse matching parallax image of the reference image according to pixel points with effective parallax and corresponding effective parallax in the reference edge binary image and the matching edge binary image, including:
determining the match in a matching binary windowThe number N of pixel points with the pixel values of both the pixel points of the edge binary image and the pixel points of the reference edge binary image in the reference binary window being 1r；
Determining the number N of pixel points with different pixel values of the pixel points of the matching edge binary image in the matching binary window and the corresponding pixel points of the reference edge binary image in the reference binary windoww；
Moving the matching binary window within a preset parallax range in the matching edge binary image according to a formula
Determining the maximum value C of Cmax；
If C is presentmaxIf the value is greater than the preset cost value, determining that the parallax value of the pixel point corresponding to the reference binary window is effective parallax, and the effective parallax is CmaxA corresponding parallax;
if C is presentmaxIf the current value is not greater than the preset cost value, determining the parallax value of the pixel point corresponding to the reference binary window as an invalid parallax;
and determining a coarse matching parallax image of the reference image according to the pixel points with effective parallax and the corresponding effective parallax in the reference edge binary image and the matching edge binary image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710823614.3A CN107680083B (en) | 2017-09-13 | 2017-09-13 | Parallax determination method and parallax determination device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710823614.3A CN107680083B (en) | 2017-09-13 | 2017-09-13 | Parallax determination method and parallax determination device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107680083A CN107680083A (en) | 2018-02-09 |
| CN107680083B true CN107680083B (en) | 2020-05-29 |
Family
ID=61136428
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710823614.3A Active CN107680083B (en) | 2017-09-13 | 2017-09-13 | Parallax determination method and parallax determination device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107680083B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108921902B (en) * | 2018-06-06 | 2020-05-08 | 贝壳技术有限公司 | Method and device for correcting structured light camera deviation |
| CN109993781B (en) * | 2019-03-28 | 2021-09-03 | 北京清微智能科技有限公司 | Parallax image generation method and system based on binocular stereo vision matching |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101625761A (en) * | 2009-08-06 | 2010-01-13 | 浙江工业大学 | Computer binocular vision matching method based on global and local algorithms |
| CN102665086A (en) * | 2012-04-26 | 2012-09-12 | 清华大学深圳研究生院 | Method for obtaining parallax by using region-based local stereo matching |
| CN102999913A (en) * | 2012-11-29 | 2013-03-27 | 清华大学深圳研究生院 | Local three-dimensional matching method based on credible point spreading |
| CN103440653A (en) * | 2013-08-27 | 2013-12-11 | 北京航空航天大学 | Binocular vision stereo matching method |
| JP2015019346A (en) * | 2013-06-12 | 2015-01-29 | 日本放送協会 | Parallax image generator |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101622344B1 (en) * | 2014-12-16 | 2016-05-19 | 경북대학교 산학협력단 | A disparity caculation method based on optimized census transform stereo matching with adaptive support weight method and system thereof |
-
2017
- 2017-09-13 CN CN201710823614.3A patent/CN107680083B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101625761A (en) * | 2009-08-06 | 2010-01-13 | 浙江工业大学 | Computer binocular vision matching method based on global and local algorithms |
| CN102665086A (en) * | 2012-04-26 | 2012-09-12 | 清华大学深圳研究生院 | Method for obtaining parallax by using region-based local stereo matching |
| CN102999913A (en) * | 2012-11-29 | 2013-03-27 | 清华大学深圳研究生院 | Local three-dimensional matching method based on credible point spreading |
| JP2015019346A (en) * | 2013-06-12 | 2015-01-29 | 日本放送協会 | Parallax image generator |
| CN103440653A (en) * | 2013-08-27 | 2013-12-11 | 北京航空航天大学 | Binocular vision stereo matching method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107680083A (en) | 2018-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107316326B (en) | Edge-based disparity map calculation method and device applied to binocular stereo vision | |
| US9070042B2 (en) | Image processing apparatus, image processing method, and program thereof | |
| JP6159298B2 (en) | Method for detecting and removing ghost artifacts in HDR image processing using multi-scale normalized cross-correlation | |
| Ansar et al. | Enhanced real-time stereo using bilateral filtering | |
| CN109448045B (en) | SLAM-based planar polygon measurement method and machine-readable storage medium | |
| CN108377374B (en) | Method and system for generating depth information related to an image | |
| CN109961417B (en) | Image processing method, image processing apparatus, and mobile apparatus control method | |
| CN111340749B (en) | Image quality detection method, device, equipment and storage medium | |
| TW200840365A (en) | Motion-blur degraded image restoration method | |
| Lo et al. | Joint trilateral filtering for depth map super-resolution | |
| JP5908844B2 (en) | Image processing apparatus and image processing method | |
| JP6976733B2 (en) | Image processing equipment, image processing methods, and programs | |
| JP2017021759A (en) | Image processor, image processing method and program | |
| US20150248745A1 (en) | Processing stereo images | |
| EP2916288A1 (en) | Camera calibration method and apparatus using a color-coded structure | |
| US20080226159A1 (en) | Method and System For Calculating Depth Information of Object in Image | |
| CN107680083B (en) | Parallax determination method and parallax determination device | |
| KR20150053438A (en) | Stereo matching system and method for generating disparity map using the same | |
| CN107403448B (en) | Cost function generation method and cost function generation device | |
| CN110800020A (en) | Image information acquisition method, image processing equipment and computer storage medium | |
| RU2580466C1 (en) | Device for recovery of depth map of scene | |
| KR101775272B1 (en) | Depth refinement method and system of sparse depth images in multi aperture camera | |
| CN107223266B (en) | Kernel approximation of fractional order differential operators for edge detection | |
| Zhang et al. | Iterative fitting after elastic registration: An efficient strategy for accurate estimation of parametric deformations | |
| CN113313645A (en) | Image processing method, image processing apparatus, terminal, and readable storage 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 |