Stereogram comparison method

Abstract

The stereogram comparison method comprises the following steps: determining a first deformation quantity of a corresponding layer of a second stereogram according to linear data of a corresponding layer of the first stereogram and linear data of a corresponding layer of the second stereogram; according to the linear data of the corresponding layer of the first stereogram, the linear data of the corresponding layer of the second stereogram and the first deformation quantity, a plurality of conversion matrixes used for converting the corresponding layer of the second stereogram into a plurality of conversion layers are generated, and a plurality of second deformation quantities of the conversion layers are smaller than the first deformation quantity; selecting a specific conversion matrix from the plurality of conversion matrixes for conversion to generate a specific conversion layer; and comparing the specific conversion layer with a corresponding layer of the first stereogram, and finding out a difference feature between the specific conversion layer and the corresponding layer of the first stereogram.

Description

Technical Field

[0001] This invention relates to a method for comparing stereoscopic images, and more particularly to a method for comparing stereoscopic images used in mechanism design. Background Technology

[0002] In the 3D structural design of printed circuit board assemblies (PCAs), customers provide structural engineers with 3D renderings of different versions of electronic components as design references. The structural engineers then modify the design based on the differences between these 3D renderings. However, PCAs involve multiple electronic components, thus requiring the consideration of numerous 3D renderings to complete the PCA's 3D structural design. Furthermore, these renderings often contain dozens of layers, each with a large number of line segments. Current technology relies on manual visual comparison to identify differences in the renderings, which is time-consuming and prone to overlooking subtle differences and erroneous modifications.

[0003] Therefore, the current technology has many shortcomings, and it is necessary to provide an improved method for stereoscopic image comparison. Summary of the Invention

[0004] This invention provides a stereoscopic image comparison method. The method includes the following steps: determining a first deformation variable of the corresponding layer of the second stereoscopic image relative to the corresponding layer of the first stereoscopic image based on multiple line type data of a corresponding layer of a first stereoscopic image and multiple line type data of a corresponding layer of a second stereoscopic image; generating multiple transformation matrices to convert the corresponding layer of the second stereoscopic image into multiple transformation layers based on the multiple line type data of the corresponding layer of the first stereoscopic image, the multiple line type data of the corresponding layer of the second stereoscopic image, and the first deformation variable, wherein multiple second deformation variables of the multiple transformation layers relative to the corresponding layer of the first stereoscopic image are smaller than the first deformation variable; selecting a specific transformation matrix from the multiple transformation matrices for transformation to generate a specific transformation layer among the multiple transformation layers; and comparing the specific transformation layer with the corresponding layer of the first stereoscopic image to identify the difference features between the specific transformation layer and the corresponding layer of the first stereoscopic image.

[0005] In some embodiments, determining a first deformation of a corresponding layer of a second stereoscopic image relative to a corresponding layer of a first stereoscopic image, based on multiple line type data of a corresponding layer of a first stereoscopic image and multiple line type data of a corresponding layer of a second stereoscopic image, includes the following steps: determining multiple line type ranges and multiple line type types of a corresponding layer of a first stereoscopic image based on multiple line type data of a corresponding layer of a first stereoscopic image; determining multiple line type ranges and multiple line type types of a corresponding layer of a second stereoscopic image based on multiple line type data of a corresponding layer of a second stereoscopic image; and determining a first deformation based on the multiple line type ranges and multiple line type types of a corresponding layer of a first stereoscopic image and the multiple line type ranges and multiple line type types of a corresponding layer of a second stereoscopic image.

[0006] In some embodiments, the stereoscopic image comparison method further includes determining a plurality of second shape variables based on a plurality of line type ranges and a plurality of line type patterns of a corresponding layer of a first stereoscopic image and a plurality of line type ranges and a plurality of line type patterns of a plurality of transformed layers.

[0007] In some embodiments, comparing a specific transformation layer with a corresponding layer of a first stereoscopic image to find the difference features between the specific transformation layer and the corresponding layer of the first stereoscopic image includes the following steps: when a first line segment of the corresponding layer of the first stereoscopic image overlaps with a second line segment of the specific transformation layer, it is determined that the second line segment is a common feature of the specific transformation layer and the corresponding layer of the first stereoscopic image.

[0008] In some embodiments, the stereoscopic image comparison method further includes the following steps: when the second stereoscopic image contains a non-corresponding layer, determining that the non-corresponding layer is a difference feature between the second stereoscopic image and the first stereoscopic image.

[0009] In some embodiments, the stereoscopic image comparison method further includes the step of aligning a first coordinate system of a first stereoscopic image with a second coordinate system of a second stereoscopic image.

[0010] In some embodiments, aligning the first coordinate system of the first stereoscopic image with the second coordinate system of the second stereoscopic image includes the step of aligning the center position of the first stereoscopic image with the center position of the second stereoscopic image.

[0011] In some embodiments, selecting a specific transformation matrix from a plurality of transformation matrices for transformation to produce a specific transformation layer among a plurality of transformation layers is performed based on a voting algorithm.

[0012] In some embodiments, a particular transformation layer has the smallest of a plurality of second shape variables.

[0013] This case proposes a stereoscopic image comparison method. The method includes the following steps: determining a first deformation variable of the corresponding layer of the second stereoscopic image relative to the corresponding layer of the first stereoscopic image based on multiple line type data of the corresponding layer of the first stereoscopic image and multiple line type data of the corresponding layer of the second stereoscopic image; generating a transformation matrix to convert the corresponding layer of the second stereoscopic image into a transformation layer based on multiple line type data of the corresponding layer of the first stereoscopic image, multiple line type data of the corresponding layer of the second stereoscopic image, and the first deformation variable, wherein the second deformation variable of the transformation layer relative to the corresponding layer of the first stereoscopic image is smaller than the first deformation variable, and further performing transformation using the transformation matrix to generate a transformation layer; and comparing the transformation layer with the corresponding layer of the first stereoscopic image to identify the difference features between the transformation layer and the corresponding layer of the first stereoscopic image. Attached Figure Description

[0014] The following paragraphs describing the implementation methods and the diagrams below will provide a better understanding of the content of this case:

[0015] Figure 1 This is a flowchart illustrating a perspective comparison method based on some embodiments of this case;

[0016] Figure 2 This is a flowchart illustrating the operation of overlay calculation and layer difference comparison according to some embodiments of this case;

[0017] Figure 3 This is a schematic diagram illustrating a method for determining differences based on some embodiments of this case;

[0018] Figure 4 A schematic diagram illustrating a display method according to some embodiments of this case; and

[0019] Figure 5 This is a flowchart illustrating the operation of overlaying and layer difference comparison according to other embodiments of this case.

[0020] Symbol explanation:

[0021] 100: Method for comparing stereoscopic images

[0022] 110: Operation

[0023] 120: Operation

[0024] 130: Operation

[0025] 140, 140A, 140B: Operation

[0026] 141A~143A: Steps

[0027] 141B~144B: Procedures

[0028] 150: Operation

[0029] 300: Difference Judgment Method

[0030] 310, 320: Layers

[0031] 400: Display Method

[0032] 410, 420: Stereoscopic Views

[0033] 430: Comparison Results

[0034] CC1~CC4: Common Features

[0035] DC1~DC4: Difference Characteristics

[0036] cL1~cL3: Corresponding layers

[0037] noncL: Non-corresponding layer

[0038] P1~P2: Line segments Detailed Implementation

[0039] The spirit of this case will be clearly explained below with diagrams and detailed description. Anyone skilled in the art can make changes and modifications based on the technology taught in this case after understanding the embodiments of this case, without departing from the spirit and scope of this case.

[0040] The terminology used herein is for the purpose of describing specific embodiments only and is not intended to limit the scope of this work. Singular forms such as “a,” “this,” “this,” “the,” and “the” as used herein also include plural forms.

[0041] The terms "include", "include", "have", "contain", etc., used in this article are all open-ended terms, meaning they include but are not limited to.

[0042] Unless otherwise specified, the terms used herein generally have their ordinary meaning in the context of the art, the subject matter, and the specific content of this case. Certain terms used to describe this case will be discussed below or elsewhere in this specification to provide additional guidance to those skilled in the art in describing the case.

[0043] Please see now Figure 1 . Figure 1The following is a flowchart illustrating a stereoscopic image comparison method 100 according to some embodiments of this invention. First, in operation 110, the two stereoscopic images to be compared are input into a stereoscopic image processing module (not shown) designed to implement the stereoscopic image comparison method 100. Specifically, the stereoscopic image processing module is stored in a storage unit of an electronic device and executed by a processor electrically connected to the storage unit. The electronic device can be a desktop computer, laptop computer, mobile phone, etc., and the storage unit can be a traditional hard disk drive, solid-state drive, etc. Next, the stereoscopic image processing module sequentially performs operations 120-150 to complete the stereoscopic image comparison method 100.

[0044] In operation 120, line data Ldata1 is generated and line data correlation operations are performed to generate line data Ldata2. In other words, the two input stereoscopic images are converted into line data Ldata1, and line data correlation operations are performed on these line data Ldata1 to generate line data Ldata2. Specifically, the two input stereoscopic images include stereoscopic image gh_1 and stereoscopic image gh_2 (not shown in the figure). Stereoscopic image gh_1 and stereoscopic image gh_2 each contain multiple layers, and each of these layers contains multiple line segments. The line data of a specific layer of stereoscopic image gh_1 or gh_2 includes the position and length of the endpoints of each of the line segments in that specific layer, where these line segments can be straight or curved. In other words, the line data of that specific layer includes the position and length of the endpoints of the straight and curved line segments in that specific layer. However, when these line segments include curved line segments, correspondingly, in addition to the positions of the two endpoints of each of the line segments in that particular layer and the length of the line segment, the line type data of that particular layer also includes the position of at least one internal point of each of the curved line segments, such as the position of the midpoint of the curved line segment.

[0045] Furthermore, the line-type data-related operations include aligning the coordinate systems of the two stereographs. Specifically, stereograph gh_1 has a coordinate system Ref_1, and stereograph gh_2 has a coordinate system Ref_2; in the line-type operations, the X-axis direction of coordinate system Ref_1 is aligned with the X-axis direction of coordinate system Ref_2, the Y-axis direction of coordinate system Ref_1 is aligned with the Y-axis direction of coordinate system Ref_2, and the Z-axis direction of coordinate system Ref_1 is aligned with the Z-axis direction of coordinate system Ref_2. In some embodiments, this can be achieved by rotating coordinate system Ref_2 using a rotation matrix.

[0046] Furthermore, in some embodiments, the center position of stereoscopic image gh_1 is aligned with the center position of stereoscopic image gh_2 to align coordinate systems Ref_1 and Ref_2, that is, the center positions of stereoscopic images gh_1 and gh_2 are at the same coordinate point. In other embodiments, the origin of coordinate system Ref_1 is aligned with the origin of coordinate system Ref_2 to align coordinate systems Ref_1 and Ref_2. Specifically, both stereoscopic images gh_1 and gh_2 have a specific point; this specific point can be set as the origin of coordinate systems Ref_1 and Ref_2. In some embodiments, the center points of stereoscopic images gh_1 and gh_2 or the origins of coordinate systems Ref_1 and Ref_2 can be aligned by moving coordinate system Ref_2 using a translation matrix.

[0047] In some embodiments, if the coordinate system converted from the 3D image is already aligned, the above-mentioned operations related to line type data can be omitted. Accordingly, line type data Ldata1 is equal to line type data Ldata2.

[0048] In operation 130, the layers of stereoscopic image gh_1 are mapped to the layers of stereoscopic image gh_2. Specifically, when designing various versions of stereoscopic images, each layer is named or labeled with a code to identify it. In operation 130, layers with the same name or code are further designated as layers to be compared. In other words, if both stereoscopic images gh_1 and gh_2 have a layer named "First Layer," then the first layer of stereoscopic image gh_1 and the first layer of stereoscopic image gh_2 are two corresponding layers. Therefore, in this document, the first layer is also referred to as the corresponding layer cL.

[0049] Next, please refer to the following: Figure 1 and Figure 2 .like Figure 1 As shown, in operation 140, overlay operations and layer difference comparisons are performed, while Figure 2 This is a flowchart illustrating operation 140A, which involves overlaying calculations and comparing layer differences, according to some embodiments of this case. In other words, operation 140A corresponds to... Figure 1 The configuration for operation 140 is as follows. Operation 140A includes steps 141A to 143A. However, the steps of operation 140A are not necessarily in the following order. Figure 3 The steps are executed sequentially. In other words, within the scope of the various embodiments of this case, steps may be appropriately added, replaced, changed in order, and / or deleted.

[0050] In step 141A, based on multiple linetype data of the corresponding layer cL of the stereoscopic image gh_1 and multiple linetype data of the corresponding layer cL of the stereoscopic image gh_2, the deformation DF1 of the corresponding layer cL of the stereoscopic image gh_2 relative to the corresponding layer cL of the stereoscopic image gh_1 is determined. Specifically, the linetype data Ldata2 processed by operation 120 includes multiple linetype data Ld2_g1 of the corresponding layer cL of the stereoscopic image gh_1 and multiple linetype data Ld2_g2 of the corresponding layer cL of the stereoscopic image gh_2.

[0051] Line data Ld2_g1 or Ld2_g2 may also include the positions of the endpoints of curved and straight line segments, and the positions of at least one interior point of each curved line segment. In some embodiments, the deformation DF1 can be determined based on the positions of the endpoints and the interior points. For example, when the positions of the endpoints and interior points are represented by position vectors, the positions of the endpoints and interior points of the line data Ld2_g1 are... The endpoints and interior points of the linear data Ld2_g2 are The positions of the endpoints and interior points of the linear data Ld2_g2 are located at... The most recent one, The positions of the endpoints and interior points of the linear data Ld2_g2 are located at... The most recent one, ... The positions of the endpoints and interior points of the linear data Ld2_g2 are located at... The most recent one. The deformation variable DF1 is a relative vector. The root mean square of the length.

[0052] Furthermore, in some embodiments, the deformation DF1 is calculated on a segment-by-segment basis. A segment can be determined by its line type range and line type shape. Specifically, based on the line type range and line type shape of segment L1 in the corresponding layer cL of the 3D image gh_1, and the line type range and line type shape of segment L2 in the corresponding layer cL of the 3D image gh_2, the deformation of segment L2 relative to segment L1 is calculated; based on the line type range and line type shape of segment L3 in the corresponding layer cL of the 3D image gh_1, and the line type range and line type shape of segment L4 in the corresponding layer cL of the 3D image gh_2, the deformation of segment L4 relative to segment L3 is calculated. This process continues until the deformation of all segments in the corresponding layer cL of the 3D image gh_2 is obtained. Then, based on the deformation of these individual segments L2, L4, ..., the deformation DF1 is estimated. In some embodiments, the deformation DF1 can be the sum of the deformation of individual segments L2, L4, ...

[0053] Furthermore, the two endpoints of a line segment determine its line type. Line types can include straight lines and / or curves. Curves can include arcs, cloud-shaped lines, etc. An arc can be determined by two endpoints and one interior point. A cloud-shaped line can be determined by two endpoints and two interior points.

[0054] In other words, we can first determine the range and type of multiple line types of the corresponding layer cL of the 3D image gh_1 based on the line type data Ld2_g1; then determine the range and type of multiple line types of the corresponding layer cL of the 3D image gh_2 based on the line type data Ld2_g2; finally, determine the deformation DF1 based on the range and type of multiple line types of the corresponding layer cL of the 3D image gh_1 and the corresponding layer cL of the 3D image gh_2.

[0055] However, it should be understood that the above-described method for calculating the deformation variable DF1 is merely an illustrative example to facilitate understanding of the stereoscopic comparison method 100 in this case, and is not intended to limit the scope of this case. Those skilled in the art can choose an appropriate algorithm to calculate the deformation variable DF1 based on the actual situation.

[0056] In step 142A, based on multiple linetype data of the corresponding layer cL of stereoscopic image gh_1, multiple linetype data of the corresponding layer cL of stereoscopic image gh_2, and deformation DF1, a transformation matrix is ​​generated to convert the corresponding layer cL of stereoscopic image gh_2 into a transformation layer TL, wherein the deformation DF2 of the transformation layer TL relative to the corresponding layer cL of stereoscopic image gh_1 is smaller than the deformation DF1. In short, the generated transformation matrix is ​​used to superimpose the corresponding layer cL of stereoscopic image gh_2 with the corresponding layer cL of stereoscopic image gh_1. In other words, the generated transformation matrix is ​​used to generate a transformation layer TL that overlaps with the corresponding layer cL of stereoscopic image gh_1 as much as possible.

[0057] In detail, the transformation matrix that satisfies the constraint that deformation variable DF2 is less than deformation variable DF1 is found through multiple loop calculations using an automated program. The calculation methods for deformation variable DF2 and DF1 are the same. The difference between the calculations of deformation variable DF2 and DF1 is that in the calculation of deformation variable DF1, the corresponding layer cL of the 3D image gh_2 is replaced with the transformation layer TL. For the sake of simplicity, the identical details of the calculations of deformation variable DF2 and DF1 will not be elaborated here.

[0058] Furthermore, the transformation matrix is ​​used to generate a transformed layer TL for subsequent layer difference comparison in step 143B.

[0059] Please refer to the following: Figure 1 and Figure 3 . Figure 3This is a schematic diagram illustrating a difference determination method 300 according to some embodiments of the present invention. In step 143A, the transformation layer TL is compared with the corresponding layer cL of the stereoscopic image gh_1 to identify the difference features between the transformation layer TL and the corresponding layer cL of the stereoscopic image gh_1. Specifically, the difference determination method 300 is used to identify the difference features between the transformation layer TL and the corresponding layer cL of the stereoscopic image gh_1. For example... Figure 3 As shown, if layers 310 and 320 overlap on line segments P1 and P2, then line segments P1 and P2 are considered common features of layers 310 and 320. Conversely, the parts of layer 310 that do not overlap with layer 320, and the parts of layer 320 that do not overlap with layer 310, are considered different features.

[0060] In other words, when line segment L_g1 of the corresponding layer cL of the stereoscopic image gh_1 overlaps with line segment L_TL of the transformation layer TL, line segment L_TL is determined to be a common feature of the transformation layer TL and the corresponding layer cL of the stereoscopic image gh_1. When line segment L_g1 of the corresponding layer cL of the stereoscopic image gh_1 does not overlap with line segment L_TL of the transformation layer TL, line segment L_TL is determined to be a difference feature of the transformation layer TL and the corresponding layer cL of the stereoscopic image gh_1. In some embodiments, when the deformation of line segment L_TL relative to line segment L_g1 is less than 0.1 times the deformation DF2, line segment L_TL is considered to overlap with line segment L_g1. When the deformation of line segment L_TL relative to line segment L_g1 is greater than or equal to 0.1 times the deformation DF2, line segment L_TL is considered not to overlap with line segment L_g1.

[0061] Furthermore, in some embodiments, one of the stereo images gh_1 and gh_2 contains a second layer, while the other does not; in this case, the second layer is called a non-corresponding layer noncL. When either stereo image gh_1 or gh_2 contains a non-corresponding layer noncL, it is determined that the non-corresponding layer noncL is a distinguishing feature between stereo images gh_1 and gh_2.

[0062] Please refer to the following: Figure 1 and Figure 4 . Figure 4 This is a schematic diagram illustrating a display method 400 according to some embodiments of the present invention. After completing operation 140A, operation 150 is performed. In operation 150, the comparison results are displayed. Specifically, in some embodiments, the comparison results are displayed using display method 400. The specific details of display method 400 are described below.

[0063] exist Figure 4In this diagram, stereoscopic images 410 and 420 correspond to stereoscopic images gh_1 and gh_2 as described above, respectively. Stereoscopic images 410 and 420 contain corresponding layers cL1 to cL3. Stereoscopic image 420 further contains a non-corresponding layer noncL1. In other words, the operations or steps described above for the corresponding layer cL can also be applied to the corresponding layers cL1 to cL3. Similarly, the operations or steps described above for the non-corresponding layer noncL can also be applied to the non-corresponding layer noncL. After operations 110 to 150, the difference between stereoscopic image 410 and stereoscopic image 420 is displayed in comparison result 430.

[0064] When a line segment in corresponding layers cL1, cL2, or cL3 is determined to be a common feature, the line segment is displayed in green, such as common features CC1 to CC4 in comparison result 430. When a line segment in corresponding layers cL1, cL2, or cL3 is determined to be a difference feature, the line segment is displayed in its original color, for example. In other words, if the line segment is contained in the stereoscopic view 420, it is displayed in red, such as difference features DC1 and DC3; if the line segment is contained in the stereoscopic view 410, it is displayed in blue, such as difference features DC2 and DC4.

[0065] Please refer to the following: Figure 1 and Figure 5 . Figure 5 This is a flowchart illustrating operation 140B, which involves overlaying operations and layer difference comparisons, according to other embodiments of this case. Similar to operation 140A, operation 140B also corresponds to... Figure 1 The operation 140 is configured in the above. In other words, the operation 140A described above can be replaced by operation 140B. In some embodiments, those skilled in the art can select operation 140A or operation 140B through a user interface, keyboard, or mouse to implement the stereoscopic comparison method 100 as needed or as required by the actual situation.

[0066] Specifically, operation 140B includes steps 141B to 144B, where step 141B is the same as step 141A in operation 140A, and therefore will not be repeated here. However, the steps of operation 140B are not necessarily in the following order. Figure 5 The steps are executed sequentially. In other words, within the scope of the various embodiments of this case, steps may be appropriately added, replaced, changed in order, and / or deleted.

[0067] In step 142B, based on the line type data of the corresponding layer cL of stereoscopic image gh_1, the line type data of the corresponding layer cL of stereoscopic image gh_2, and the deformation DF1, multiple transformation matrices are generated to transform the corresponding layer cL of stereoscopic image gh_2 into multiple transformation layers TL, wherein the deformation DF2 of the multiple transformation layers TL relative to the corresponding layer cL of stereoscopic image gh_1 is smaller than the deformation DF1. Furthermore, there is a one-to-one correspondence between the transformation matrix, the transformation layer TL, and the deformation DF2.

[0068] Compared to step 142A, the difference between step 142B and step 142A is that step 142B generates multiple transformation matrices instead of a single transformation matrix. Apart from this difference, the other descriptions of step 142A are applicable to step 141A and will not be repeated here. This difference implies that the constraint that the transformation matrix for which deformation DF2 is less than deformation DF1 is not unique is not satisfied. The transformation matrix has multiple possibilities. In this case, it is necessary to select the most suitable transformation matrix for the transformation through certain selection rules or more constraints.

[0069] Therefore, in step 143B, a specific transformation matrix is ​​selected from multiple transformation matrices for transformation to produce a specific transformation layer among multiple transformation layers. Specifically, in some embodiments, the transformation matrix with the smallest deformation DF2 is used as the specific transformation matrix for transformation. In some embodiments, step 143B is performed based on a voting algorithm. In other words, multiple transformations are scored according to the principles set by the voting algorithm, and the transformation matrix with the highest score is selected as the specific transformation matrix for transformation.

[0070] In step 144B, the specific transformation layer is compared with the corresponding layer cL of the stereoscopic image gh_1 to identify the differences between the specific transformation layer and the corresponding layer of the stereoscopic image. Compared to step 143A, the difference between step 144B and step 143A lies in using a specific transformation layer generated by a selected transformation matrix for layer difference comparison. Furthermore, it should be understood that, apart from this difference, the description of the transformation layer TL in step 143A applies to the specific transformation layer in step 144B.

[0071] In summary, this case provides a standardized method for comparing stereoscopic images, which can save a lot of manpower and bring convenience to mechanical engineers in modifying mechanical designs, thus making up for the shortcomings of previous technologies.

[0072] Although this case has been disclosed above with detailed embodiments, it does not exclude other possible implementations. Therefore, the scope of protection of this case shall be determined by the appended claims, and not by the limitations of the foregoing embodiments.

Claims

1. A method for comparing stereoscopic images, characterized in that, Include: Based on a plurality of line type data of a corresponding layer of a first stereoscopic image and a plurality of line type data of a corresponding layer of a second stereoscopic image, a first deformation of the corresponding layer of the second stereoscopic image relative to the corresponding layer of the first stereoscopic image is determined; Based on the plurality of line type data of the corresponding layer of the first stereoscopic image, the plurality of line type data of the corresponding layer of the second stereoscopic image, and the first deformation, a plurality of transformation matrices are generated to convert the corresponding layer of the second stereoscopic image into a plurality of transformation layers, wherein the plurality of transformation layers have a plurality of second deformations relative to the corresponding layer of the first stereoscopic image that are smaller than the first deformation. Select a specific transformation matrix from the plurality of transformation matrices and perform the transformation to generate a specific transformation layer from the plurality of transformation layers; as well as The specific transformation layer is compared with the corresponding layer of the first stereoscopic image to identify a difference feature between the specific transformation layer and the corresponding layer of the first stereoscopic image.

2. The stereoscopic image comparison method as described in claim 1, characterized in that, The first deformation of the corresponding layer of the second stereoscopic image relative to the corresponding layer of the first stereoscopic image, based on the plurality of line type data of the corresponding layer of the first stereoscopic image and the plurality of line type data of the corresponding layer of the second stereoscopic image, includes: Based on the plurality of line type data of the corresponding layer of the first stereoscopic image, determine the plurality of line type ranges and the plurality of line type types of the corresponding layer of the first stereoscopic image; Based on the plurality of linetype data of the corresponding layer of the second stereoscopic image, determine the plurality of linetype ranges and plurality of linetype types of the corresponding layer of the second stereoscopic image; and The first deformation is determined based on the plurality of line type ranges and line type patterns of the corresponding layer of the first stereoscopic image and the plurality of line type ranges and line type patterns of the corresponding layer of the second stereoscopic image.

3. The stereoscopic image comparison method as described in claim 2, characterized in that, It also includes: The plurality of second deformation variables are determined based on the plurality of line type ranges and line type patterns of the corresponding layers of the first stereoscopic image and the plurality of line type ranges and line type patterns of the plurality of transformation layers.

4. The stereoscopic image comparison method as described in claim 1, characterized in that, The comparison of the specific transformation layer with the corresponding layer of the first stereoscopic image to identify the difference features between the specific transformation layer and the corresponding layer of the first stereoscopic image includes: When a first line segment of the corresponding layer of the first stereoscopic image overlaps with a second line segment of the specific transformation layer, it is determined that the second line segment is a common feature of the specific transformation layer and the corresponding layer of the first stereoscopic image.

5. The stereoscopic image comparison method as described in claim 1, characterized in that, It also includes: When the second stereoscopic image contains a non-corresponding layer, it is determined that the non-corresponding layer is the difference feature between the second stereoscopic image and the first stereoscopic image.

6. The stereoscopic image comparison method as described in claim 1, characterized in that, It also includes: Align a first coordinate system of the first stereoscopic image with a second coordinate system of the second stereoscopic image.

7. The stereoscopic image comparison method as described in claim 6, characterized in that, Aligning the first coordinate system of the first stereoscopic image with the second coordinate system of the second stereoscopic image includes: Align one center position of the first stereoscopic image with one center position of the second stereoscopic image.

8. The stereoscopic image comparison method as described in claim 1, characterized in that, The selection of a specific transformation matrix from the plurality of transformation matrices to generate a specific transformation layer from the plurality of transformation layers is performed based on a voting algorithm.

9. The stereoscopic image comparison method as described in claim 1, characterized in that, The particular transformation layer has the smallest of the plurality of second shape variables.

10. A method for comparing stereoscopic images, characterized in that, Include: Based on a plurality of line type data of a corresponding layer of a first stereoscopic image and a plurality of line type data of a corresponding layer of a second stereoscopic image, a first deformation of the corresponding layer of the second stereoscopic image relative to the corresponding layer of the first stereoscopic image is determined; Based on the plurality of line type data of the corresponding layer of the first stereoscopic image, the plurality of line type data of the corresponding layer of the second stereoscopic image, and the first deformation, a transformation matrix is ​​generated to convert the corresponding layer of the second stereoscopic image into a transformation layer, wherein a second deformation of the transformation layer relative to the corresponding layer of the first stereoscopic image is smaller than the first deformation, and the transformation is further performed using the transformation matrix to generate the transformation layer; as well as The transformed layer is compared with the corresponding layer of the first stereoscopic image to identify a difference feature between the transformed layer and the corresponding layer of the first stereoscopic image.

Images