Image recognition-based three-dimensional simulation method and auxiliary device

Abstract

The application discloses a kind of based on image recognition three-dimensional simulation simulation method and auxiliary device, by the three-dimensional image is divided into several regional division after layering, obtain the image of multiple faces and be spliced, then select small size local filling edge detail part, ensure that each contour is completely covered, reduce the gap between line segment after the model is established in identification image, reduce error, avoid the part of unclear broken line in model and generate error information.The sweep plate in auxiliary device can limit and avoid the side wall of movable piece from being excessively squeezed, at this time, the side wall of vertical frame will be pulled outwards, so that the inside of vertical frame changes from vertical state to inclined angle and gradually increases, which can drive the scanner to scan the bottom side wall of target object at multiple angles.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image recognition, in particular to a three-dimensional simulation method based on image recognition and an auxiliary device. BACKGROUND

[0002] When a three-dimensional target object is scanned, there are concave and convex parts of small details everywhere. The parts of small details cannot be constructed into a complete model at one time. No matter from which angle the operation is performed, there will always be parts of inaccuracy in connection or disconnection of the interface in the recessed parts, so that the finally constructed model is always incomplete. In the process of image recognition, errors are easily generated in the process of aligning the entire three-dimensional body with the target object, and new errors occur. SUMMARY

[0003] To solve the problem that the three-dimensional object imaging is easy to miss the details and generate errors, the purpose of the present application is to provide a three-dimensional simulation method based on image recognition and an auxiliary device, which realizes split scanning, avoids inaccurate connection or disconnection of line segments, and accurately images the three-dimensional object.

[0004] The purpose of the present application is achieved by the following technical solutions:

[0005] A three-dimensional simulation method based on image recognition, comprising the following steps:

[0006] S1, constructing a digital model, first scanning three-dimensional topographic information of a target object, using image segmentation technology to extract two-dimensional contour information of the object in the image, fusing the three-dimensional information and the two-dimensional contour information to obtain a digital model of the object, the state presented by the digital model being the same as the state of the outer line of the target object;

[0007] S2, dividing the image into several regional modules, into two levels of outer frame and inner frame, and then refining and layering to obtain two-dimensional images of each face, using a morphological method to perform defect repair processing on the two-dimensional images, extracting the peripheral contour of the target object, and supplementing the two-dimensional images of each face of the object to the peripheral contour;

[0008] S3, using a structural element to perform morphological transformation on the image, when selecting the structural element, small size structural elements have better protection of edge details and better noise removal ability, and large size structural elements have stronger noise removal ability;

[0009] S4, combining large size and small size structural elements for image establishment, and then merging the two established size structural elements;

[0010] S5, the parameterization of the data points ensures that the curved surface construction three-dimensional simulation accuracy, according to the data points in the digital model according to its chord length, in the actual image display model reaction data point chord length distribution, along the edge of the image point;

[0011] S6, first draw a large size contour, then add small contour point size in the range of large size scanning, fill the missing holes, the missing parts are filled, and the parts are aligned with the original large size model.

[0012] S7, the outside wall of the supplemented alignment part is re-supplemented, which ensures that each contour is fully covered, reduces the gap between the line segments in the identified image after the model is established, and reduces the error.

[0013] An image recognition three-dimensional simulation auxiliary device based on, including movable frame, the inner side wall of movable frame is connected through scanning frame mechanism, the inside of scanning frame mechanism includes limit ring, movable piece and telescopic frame, the inner side wall of limit ring is connected through the outer side wall of movable piece, the inner side of movable piece is connected through the outer side wall of telescopic frame, the side wall of scanning frame mechanism is connected through slide rod, the side wall of slide rod is connected through scanning plate, the bottom of movable frame is connected through swing mechanism, the inside of swing mechanism includes support seat, arc plate and movable shell, the inner side wall of support seat is connected through the outer end of arc plate, the inner side wall of arc plate is connected through the outer side wall of movable shell, the inner side wall of scanning frame mechanism is connected through vertical frame, the side wall of swing mechanism is connected through elastic cable, the side wall of elastic cable is connected through movable ball.

[0014] Further, the side walls of the movable piece are slidably connected, the side walls of the movable piece are tightly buckled, and the movable piece can also be pulled outward.

[0015] Further, the side wall of the scanning plate is connected through and slidably connected to the side wall of the scanning frame mechanism, and the scanning plate can move outward with the stretching of the side wall of the scanning frame mechanism.

[0016] Further, the inner side wall of the movable shell is slidably connected to the outer side wall of the movable ball, and after the rotation between the side wall of the support seat and the side wall of the movable shell, the side wall of the elastic cable can be pulled, and the side wall of the elastic cable can drive the rotation of the side wall of the movable ball.

[0017] Further, the outer side wall of the vertical frame is connected through and movably connected to the inner side wall of the telescopic frame, and when the side wall of the telescopic frame is telescoped, the side wall of the vertical frame can be tilted and rotated.

[0018] Further, the bottom of the scanning frame mechanism is connected through the inner side wall top of the swing mechanism, and when the side wall of the scanning frame mechanism moves, the side wall of the swing mechanism can be inclined and pressed.

[0019] Further, the side wall of the sweep plate is connected to the side wall gap of the limiting ring through and slidingly, and the sweep plate can limit the side wall of the movable piece from being excessively pressed.

[0020] The present application has the following advantages:

[0021] 1. The three-dimensional simulation method based on image recognition divides the three-dimensional image into several regions, performs layering after division, obtains multiple images for splicing, selects small-size local filling edge detail parts, ensures that each contour is fully covered, reduces the gap between line segments after the model is established, reduces errors, and avoids generating incorrect information due to unclear broken lines in the model.

[0022] 2. The three-dimensional simulation method based on image recognition pulls the movable piece outward through the side wall of the limiting ring, pulls the side wall of the movable piece apart from the mutual clamping state, lengthens the side wall of the limiting ring outward, limits the side wall of the movable piece from being excessively pressed, and at this time, pulls the side wall of the vertical frame outward, changes the inside of the vertical frame from a vertical state to an inclined angle and gradually increases the angle, and can drive the scanner to scan the bottom side wall of the target object at multiple angles.

[0023] 3. The three-dimensional simulation method based on image recognition, the top of the swing mechanism is pressed by the movable frame to be inclined, the side wall of the support seat is inclined and rotated outward, the side wall of the support seat and the side wall of the movable shell are rotated, the side wall of the elastic cord is pulled, the side wall of the elastic cord drives the side wall of the movable ball to rotate, thereby uniformly pushing the swing mechanism, conveniently limiting the uniform expansion and contraction of the side wall of the swing mechanism, and enabling the scanner to move up and down at the same height on the outside wall of the target object. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The present application is a method flowchart;

[0025] Figure 2 The present application is a structure connection overall schematic diagram;

[0026] Figure 3 The present application is a top view of the internal related structure of the sweep frame mechanism;

[0027] Figure 4 The present application is a connection diagram of the internal related structure of the swing mechanism.

[0028] In the figure: 1, movable frame; 2, sweep frame mechanism; 211, limiting ring; 212, movable piece; 213, telescopic frame; 3, slide rod; 4, sweep plate; 5, swing mechanism; 511, support seat; 512, arc plate; 513, movable shell; 6, vertical frame; 7, movable ball; 8, elastic cord. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of protection of the present application.

[0030] A three-dimensional simulation method based on image recognition, please refer to Figure 1 , comprising the following steps:

[0031] S1, constructing a digital model, first scanning the three-dimensional topographic information of the target object, using image segmentation technology to extract the two-dimensional contour information of the object in the image, after fusing the three-dimensional information and the two-dimensional contour information, obtaining the digital model of the object, the state presented by the digital model is the same as the state of the outer line of the target object;

[0032] S2, the image is divided into several regional modules, divided into two levels of outer frame and inner frame, and then refined and layered to obtain two-dimensional images of each face, using morphological method to process the defects of the two-dimensional images, extracting the peripheral contour of the target object, and supplementing the two-dimensional images of each face of the object to the peripheral contour;

[0033] S3, using a structural element to perform morphological transformation on the image, when selecting the structural element, for different sizes and shapes, small size structural element has better noise removal ability and better protection of edge details, large size structural element has stronger noise removal ability;

[0034] S4, for poor protection of edge details, combine large size and small size structural elements, respectively, to establish the image, and then merge the two established size structural elements;

[0035] S5, the parameterization of the data points ensures the accuracy of the three-dimensional simulation of the curved surface, according to the data points in the digital model, the actual image shows the distribution of the data point chord length, and the edge in the image is described;

[0036] S6, first draw the large size contour, then add small contour points within the large size range, fill in the missing gaps, and then align the filled parts with the original large size model.

[0037] S7, re-supplement scanning outside the wall of the supplemented alignment site, ensure that each profile is fully covered, reduce the gap between the line segments in the identified image after the model is established, reduce the error.

[0038] Please refer to Figures 2-4 An image recognition-based three-dimensional simulation auxiliary device, comprising a movable frame 1, a scanning frame mechanism 2 is connected through the inner side wall of the movable frame 1, the inside of the scanning frame mechanism 2 comprises a limiting ring 211, a movable piece 212 and a telescopic frame 213, the inner side wall of the limiting ring 211 is connected through the outer side wall of the movable piece 212, the inner side of the movable piece 212 is connected through the outer side wall of the telescopic frame 213, a sliding rod 3 is connected through the side wall of the scanning frame mechanism 2, a scanning plate 4 is connected through the side wall of the sliding rod 3, a swing mechanism 5 is connected through the bottom end of the movable frame 1, the inside of the swing mechanism 5 comprises a supporting seat 511, an arc plate 512 and a movable shell 513, the inner side wall of the supporting seat 511 is connected through the outer end of the arc plate 512, the inner side wall of the arc plate 512 is connected through and slides on the outer side wall of the movable shell 513, a vertical frame 6 is connected through the inner side wall of the scanning frame mechanism 2, an elastic cord 8 is connected through the side wall of the swing mechanism 5, and a movable ball 7 is connected through the side wall of the elastic cord 8.

[0039] Wherein, the side walls of the movable piece 212 are connected through sliding, the side walls of the movable piece 212 are buckled to each other, and can also be pulled outward.

[0040] Wherein, the side wall of the scanning plate 4 is connected through and slides on the side wall of the scanning frame mechanism 2, the scanning plate 4 can move outward with the stretching of the side wall of the scanning frame mechanism 2.

[0041] Wherein, the inner side wall of the movable shell 513 is connected through and slides on the outer side wall of the movable ball 7, the side wall of the elastic cord 8 can be pulled after the side wall of the supporting seat 511 and the side wall of the movable shell 513 are rotated, and the side wall of the elastic cord 8 can drive the side wall of the movable ball 7 to rotate.

[0042] Wherein, the outer side wall of the vertical frame 6 is connected through and moves on the inner side wall of the telescopic frame 213, and the side wall of the telescopic frame 213 can drive the side wall of the vertical frame 6 to tilt and rotate when the telescopic frame 213 is telescoped.

[0043] Wherein, the bottom end of the scanning frame mechanism 2 is connected through the inner side wall top end of the swing mechanism 5, and the side wall of the scanning frame mechanism 2 can drive the side wall of the swing mechanism 5 to tilt and press when the side wall of the scanning frame mechanism 2 moves.

[0044] Wherein, the side wall of the scanning plate 4 is connected through and slides on the side wall gap of the limiting ring 211, and the scanning plate 4 can limit and avoid excessive extrusion of the side wall of the movable piece 212.

[0045] In use, such as Figure 2 And Figure 3As shown, the target object is at the inner ring side wall of the swing mechanism 5, the side wall of the operating movable frame 1 is stretched up and down to the high and low positions, because the inner ring side wall of the sweep frame mechanism 2 is at the outer side of the upper end of the target object, and the side wall of the movable frame 1 is around the outer ring side wall of the sweep frame mechanism 2, when the side wall of the movable frame 1 is stretched outwards, it can frame and cover around the outer side of the target object, when the side wall of the movable frame 1 is pressed downwards, the bottom end of the side wall of the movable frame 1 increases the angle of inclination outward, pulling the side wall of the sweep frame mechanism 2, pushing the side wall of the limiting ring 211 outward, the side wall of the limiting ring 211 pulls the movable piece 212, after the side wall of the movable piece 212 is pulled away from the state of mutual buckling, the side wall of the limiting ring 211 is pulled outward, the sweep plate 4 can limit the excessive extrusion of the side wall of the movable piece 212, at this time, it will pull the side wall of the vertical frame 6 outward, so that the inner side of the vertical frame 6 changes from the vertical state to the inclined angle and gradually increases, which can drive the scanner to scan the bottom side wall of the target object at multiple angles.

[0046] As shown in Figure 4 The top end of the swing mechanism 5 is pressed by the movable frame 1 to be inclined, the side wall of the supporting seat 511 is inclined and rotated outward, and after the rotation between the side wall of the supporting seat 511 and the side wall of the movable shell 513, the side wall of the elastic cable 8 is pulled, the side wall of the elastic cable 8 drives the side wall of the movable ball 7 to rotate, thereby uniformly pushing the swing mechanism 5, conveniently limiting the uniform stretching of the side wall of the swing mechanism 5, and enabling the sweep frame mechanism 2 to move up and down on the outer side wall of the target object at the same height.

[0047] Although embodiments of the present application have been shown and described, it is to be understood that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An auxiliary device based on image recognition and three-dimensional simulation, characterized in that: The device includes a movable frame (1), with a sweeping mechanism (2) penetrating its inner wall. The sweeping mechanism (2) includes a limiting ring (211), a movable piece (212), and a telescopic frame (213). The inner wall of the limiting ring (211) is connected to the outer wall of the movable piece (212), and the inner side of the movable piece (212) is connected to the outer wall of the telescopic frame (213). A sliding rod (3) is penetrating its side wall, and a sweeping plate (4) is penetrating its side wall. The bottom end of the sweeping frame mechanism (2) is connected to a swinging mechanism (5). The swinging mechanism (5) includes a support base (511), an arc plate (512), and a movable shell (513). The inner side wall of the support base (511) is connected to the outer end of the arc plate (512), and the inner side wall of the arc plate (512) is connected to the outer side wall of the movable shell (513). The inner side wall of the sweeping frame mechanism (2) is connected to a vertical frame (6), the side wall of the swinging mechanism (5) is connected to an elastic cable (8), and the side wall of the elastic cable (8) is connected to a movable ball (7). In use, the target object is located on the inner side wall of the swing mechanism (5). The side wall of the movable frame (1) is moved up and down to adjust its height. Since the inner side wall of the sweeping frame mechanism (2) is located on the upper outer side of the target object, and the side wall of the movable frame (1) is wrapped around and passes through the outer side wall of the sweeping frame mechanism (2), when the side wall of the movable frame (1) opens outward, it frames and covers the outer perimeter of the target object. When the side wall of the movable frame (1) moves downward and presses down, the bottom end of the side wall of the movable frame (1) tilts outward at an increased angle, pulling the side wall of the sweeping frame mechanism (2) and pushing the side wall of the limiting ring (211) outward. The wall pulls the movable piece (212) outward. After the side walls of the movable piece (212) are pulled apart from the interlocked state, the side wall of the limiting ring (211) is stretched outward. The scanning plate (4) restricts and avoids excessive compression of the side wall of the movable piece (212). At this time, the side wall of the vertical frame (6) will be pulled outward, so that the inner side of the vertical frame (6) changes from a vertical state to an inclined angle and gradually increases thereafter. The scanner can be driven to scan the bottom side wall of the target object completely from multiple angles. The side wall of the scanning plate (4) is connected to the side wall of the scanning frame mechanism (2) through and sliding. The scanning plate (4) can move outward as the side wall of the scanning frame mechanism (2) is stretched. The top of the swing mechanism (5) will be pressed to tilt by the movable frame (1), and the side wall of the support seat (511) will tilt and rotate outward. After the side wall of the support seat (511) and the side wall of the movable shell (513) rotate, the side wall of the elastic cable (8) can be pulled. The side wall of the elastic cable (8) can drive the side wall of the movable ball (7) to rotate, thereby uniformly pushing the swing mechanism (5), which facilitates the uniform extension and contraction of the side wall of the swing mechanism (5), and can move up and down at the same height on the outer side wall of the target object.

2. The auxiliary device for three-dimensional simulation based on image recognition according to claim 1, characterized in that: The sidewalls of the movable piece (212) are slidably connected.

3. The auxiliary device for three-dimensional simulation based on image recognition according to claim 1, characterized in that: The inner wall of the movable shell (513) is slidably connected to the outer wall of the movable ball (7).

4. The auxiliary device for three-dimensional simulation based on image recognition according to claim 1, characterized in that: The outer wall of the vertical frame (6) is connected through and movably to the inner wall of the telescopic frame (213).

5. The auxiliary device for three-dimensional simulation based on image recognition according to claim 1, characterized in that: The bottom end of the sweeping mechanism (2) is connected to the top of the inner wall of the swing mechanism (5).

6. The auxiliary device for three-dimensional simulation based on image recognition according to claim 1, characterized in that: The sidewall of the sweeping plate (4) is slidably connected to the sidewall gap of the limiting ring (211).

7. A method for using the auxiliary device based on image recognition three-dimensional simulation as described in any one of claims 1-6, characterized in that, Includes the following steps: S1. Constructing a digital model: First, scan the three-dimensional shape information of the target object. Use image segmentation technology to extract the two-dimensional contour information of the object in the image. After fusing the three-dimensional information with the two-dimensional contour information, a digital model of the object is obtained. The state presented by the digital model is the same as the outer line state of the target object. S2. Divide the image into several regional modules, into two levels: outer frame and inner frame. Then refine and layer them to obtain two-dimensional images of each face. Use morphological methods to repair defects in the two-dimensional images. After extracting the outer contour of the target object, fill the two-dimensional images of each face of the object onto the outer contour. S3. Use structural elements to perform morphological transformations on the image. When selecting structural elements, depending on their size and shape, smaller structural elements have weaker noise reduction capabilities but better protection of edge details, while larger structural elements have stronger noise reduction capabilities. S4. Poor protection of edge details. Combine large and small structural elements, build the image separately, and then merge the structural elements of the two sizes. S5. When parameterizing data points to ensure the accuracy of 3D simulation of surface construction, the data points in the digital model are plotted along the edges of the image to reflect the distribution of the chord lengths of the data points in the actual image, according to their own chord lengths. S6. First, draw the outline of the large size. Then, add and scan the point size of the small outline within the large size range to fill in the missing gaps. After filling in the missing parts, align the filled parts with the original large size model. S7. Rescan the outer wall of the aligned area to ensure that each contour is fully covered, reduce gaps between line segments in the recognition image after model building, and reduce errors.

Images