Measurement and identification system for 3D objects

Abstract

A measurement and identification system for 3D objects, comprising: a measurement platform (1), at least one image capture unit (2), and an image processing unit (3). The measurement platform (1) is used for bearing a 3D object to be measured; the image capture unit (2) is provided on the measurement platform (1) to capture an image of said 3D object; and upon receiving the image of said 3D object, the image processing unit (3) can directly measure the actual dimension of the 3D object in the image, and can identify whether the appearance of the 3D object in the image is crooked and deformed, determine whether the color of the object and accessory positions are correct, and determine whether any accessory is absent, so as to overcome the limitations of measurement with human eyes and subjective identification errors, thereby improving outgoing quality of manufactured objects.

Description

Measurement and recognition system for 3D objects Technical Field

[0001] The present invention relates to an image recognition system, and more particularly to a measurement and recognition system that can capture images of door and window building materials to determine whether the 3D object size of the door and window building materials is correct and whether the object accessories are complete. Background Art

[0002] Currently, with the exception of aluminum, which must comply with the GB / T5237-2000 "Aluminum Alloy Building Profiles" specification, other building materials often undergo complex processes such as cutting, stamping and drilling, assembly, commissioning, and gluing. Furthermore, different building material styles require different processes due to the different accessories such as concealed locks, deadbolts, rear buttons, mortise locks, and ceiling bolts. Therefore, to ensure that door and window materials meet customer needs, many building material manufacturers not only inspect the appearance and dimensions of each frame upon completion, but also review all accessories. Only after the dimensions and accessories have been inspected and verified are they allowed to be stored or shipped.

[0003] However, the current door and window building materials manufacturing industry relies on manual measurement for building material dimensions and visual verification of the correctness of building material accessories. However, human visual inspection is not only limited but also prone to subjective errors. Furthermore, due to insufficient lighting in the inspection environment, the captured image is too dark, making it impossible to identify or inaccurate. Furthermore, due to errors such as the inspector's physiological state or improper operation of the equipment, errors in inspections are common.

[0004] Summary of the Invention

[0005] The technical solution adopted by the present invention is: a measurement and identification system for 3D objects, which includes: a measurement platform, at least one image capture unit and an image processing unit, wherein the measurement platform is used to support a 3D object to be measured, the image capture units are arranged on the measurement platform and electrically connected to the image processing unit, the image capture unit captures information and images of the 3D object and transmits them to the image processing unit, and after receiving the 3D object image and information, the image processing unit measures and identifies the actual information of the 3D object in the image.

[0006] The beneficial effects of the present invention are as follows: The main purpose of the present invention is to provide a system for 3D object measurement and recognition to overcome the limitations of human eye measurement and subjective cognitive errors in recognition, thereby improving the accuracy of building material detection in doors and windows.

[0007] Another main purpose of the present invention is to provide a 3D object measurement and recognition system to overcome the problem that the captured images of building materials such as doors and windows are often too dark due to insufficient light in the inspection environment, resulting in the inability to identify or insufficient recognition accuracy.

[0008] The present invention has the following advantages: the image processing unit stores standard information about building materials for doors and windows, such as size, appearance, shape, color, accessories, and accessory positions. The scanning device of the image capture unit scans the 3D object recognition unit to obtain the 3D object information, and the image capture device captures the image of the 3D object and transmits the 3D object information and image to the image processing unit. After capturing the 3D object image and information, the image processing unit can directly measure the actual size of the 3D object and then identify whether the appearance of the 3D object in the image is skewed or deformed, whether the color is correct, whether the accessories are complete, and whether the accessories are misplaced. Therefore, by measuring and identifying the image as described above, it is possible to eliminate the problems of incorrect detection or identification of the object to be tested caused by the physiological state of the daily inspector or improper operation of the instrument. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG1 is a perspective schematic diagram of the present invention.

[0010] FIG2 is a front view of the measuring platform of the present invention.

[0011] FIG3 is a schematic diagram of a measuring platform supporting a building to be measured according to an embodiment of the present invention.

[0012] FIG. 4 is a block diagram illustrating the flow of an image processing unit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0013] Please refer to Figures 1 to 3, which are a preferred embodiment of a measurement and recognition system for 3D objects of the present invention, comprising: a measurement platform (1), at least one image capture unit (2) and an image processing unit (3), and the image capture unit (2) comprises a scanning device (not shown in the figure) and an image capture device (not shown in the figure); wherein the measurement platform (1) has at least an upper platform (12) and a lower platform (11), and the lower platform (11) is provided with at least one bottom light source (16), and the upper platform (12) is provided with at least one top light source (17), and the lower platform (11) is provided with a transparent substrate (14) for holding a 3D object (A) to be measured, and the image capture units (2) are provided on the upper platform (12) and electrically connected to the image processing unit (3), and the scanning device of the image capture unit (2) is an identification unit for scanning the 3D object (A), such as a QR code. The code is used to obtain information about the 3D object (A), and the image capture device captures the image of the 3D object (A), such as appearance, shape, color, length, width, thickness, etc. The image capture unit captures the image of the 3D object (A) and converts it into an image signal, which is then transmitted to the image processing unit (3). After receiving the image signal of the 3D object (A), the image processing unit (3) measures the size of the 3D object (A) in the image and can also identify at least one of the color of the 3D object (A) in the image, whether the appearance is skewed or deformed, whether the accessories are complete, and whether the accessories are correctly positioned. In this embodiment, the transparent substrate (14) is a tempered glass, but is not limited to the above.

[0014] Please refer to Figures 1 to 3, in which the upper platform (12) is mounted on the lower platform (11), and the upper platform (12) has at least one X-axis slide rail (121), one Y-axis slide rail (122) and one Z-axis slide rail (123), and the image capture unit (2) is arranged on the upper platform (12), that is, the image capture unit (2) can be arranged on at least one of the X-axis slide rail (121), the Y-axis slide rail (122) or the Z-axis slide rail (123). In this embodiment, the image capture unit (2) is provided on each of the X-axis slide rail (121), the Y-axis slide rail (122) and the Z-axis slide rail (123) to obtain the image of the length, width and height of the 3D object (A), and the image capture device of the image capture unit (2) can be A high-resolution camera, such as a CMOS industrial camera, or a video camera, however, regardless of the type of image capture device, the image capture device has at least a lens and a photosensitive coupling component, wherein the lens is used to capture the image of the 3D object (A), and the photosensitive coupling component is connected to the image processing unit (3) by wireless or wired signal connection. Therefore, after the lens captures the image of the 3D object (A), the photosensitive coupling component converts the captured image into an image signal, and then transmits it to the image processing unit (3) for image processing. After receiving the image signal, the image processing unit (3) measures the size of the 3D object (A) in the image signal and identifies whether the object's accessories are complete.

[0015] Please refer to FIG. 1 and FIG. 4 again. In this embodiment, the image processing unit (3) is a computer. In addition to having a built-in image processing software (31), the image processing unit (3) stores standard information of building materials of doors and windows, such as actual information of size, appearance shape, color, accessories and accessory positions. Before measuring and identifying the image of the 3D object (A), the scanning device of the image capture unit (2) scans the recognition unit of the 3D object (A) to obtain the information of the 3D object (A), and the image capture unit (2) scans the recognition unit of the 3D object (A) to obtain the information of the 3D object (A). The image capturing device of the unit (2) captures the image of the 3D object (A) and transmits the information and image of the 3D object (A) to the image processing unit (3). Therefore, after the image processing unit (3) obtains the information and image of the 3D object (A), it first analyzes the information and image of the 3D object (A) through an image comparison module (311) of the image processing software (31), compares the area of ​​the 3D object (A) image that needs to be measured, and then uses an image comparison module (311) of the image processing software (31) to analyze the information and image of the 3D object (A). The edge detection module (312) obtains the pixel length and width of the area to be measured in the image of the 3D object (A), and then compares the length and width with the standard length and width dimensions of the building materials of the door and window that have been pre-stored and calculates to obtain the actual length and width of the 3D object area in the image of the 3D object (A); Furthermore, the image processing unit (3) of the present invention further includes an image recognition system (32), such as AI image recognition, and the image recognition system can directly measure and / or recognize the captured image. Analysis, and the image recognition system must first define the actual image data of the standard size, color, number of accessories and their position of the building materials of doors and windows, and then set the parameters of the color of the 3D object (A) in the image, whether the appearance is skewed or deformed, whether the accessories are complete, and whether the accessories are correctly positioned as the basis for recognition, so that this implementation can be flexible and fast during recognition, reducing the situation of insufficient recognition rate; furthermore, the present invention can also measure the size of building materials of doors and windows through calculation software with similar functions, and is not limited to the embodiments disclosed above.

[0016] Furthermore, the image processing software (31) compares the image of the 3D object (A) with the standard image of the building materials of doors and windows through an analysis module (313) to detect whether the appearance of the 3D object in the image is skewed or deformed (commonly known as bulging), and to identify the color and accessories of the building materials of doors and windows. The image comparison performed by the analysis module (313) is used to search for the same accessories in the same position in the image of the 3D object (A), and then analyze whether the position of the accessories is correct. When the appearance is deformed, the color is incorrect, the accessories are missing, or the position of the accessories is incorrect, the image processing unit (3) will automatically generate an abnormal warning notification. In addition, the present invention can also be completed by computing software with similar functions and is not limited to the embodiments disclosed in the present invention.

[0017] Please refer to Figures 1 to 3 again. The upper platform (12) of this embodiment is provided with multiple groups of top light sources (17). Therefore, in this embodiment, the light emitted by the top light source (17) and the bottom light source (16) can increase the brightness of the 3D object (A) on the transparent substrate (14), so that the image of the 3D object (A) captured by the image capture unit (2) will not be too dark due to insufficient light, thereby overcoming the problem that the conventional technology detection environment is often too dark due to insufficient light, resulting in the captured image being unable to be identified or the identification accuracy being insufficient; in this embodiment, the bottom light source (16) and the top light source (17) can be a layer of board lights, but this is limited to the aforementioned embodiment. Any lamp that can provide a sufficiently bright light source on the transparent substrate (14) falls within the scope of application of this embodiment.

[0018] The above description is only used to explain the preferred embodiments of the present invention and is not intended to limit the present invention in any form. Therefore, any modifications or changes made to the present invention under the same inventive spirit should still be included in the scope of protection intended by the present invention.

Claims

1. A system for measuring and identifying 3D objects, characterized by: It includes: A measurement platform, at least one image capture unit, and an image processing unit, wherein the measurement platform is used to support a 3D object to be measured. The image capture units are arranged on the measurement platform and electrically connected to the image processing unit. The image capture unit captures information and images of the 3D object and transmits them to the image processing unit. After receiving the 3D object image and information, the image processing unit measures and identifies the actual information of the 3D object in the image.

2. The 3D object measurement and recognition system according to claim 1, wherein: After receiving the 3D object image and information, the image processing unit measures the size of the 3D object in the image and identifies at least one of the color, whether the appearance is skewed or deformed, whether the accessories are complete, and whether the accessories are positioned correctly.

3. The 3D object measurement and recognition system according to claim 1, wherein: The measuring platform at least has an upper platform and a lower platform, and a transparent substrate is provided on the lower platform for the 3D object to be placed.

4. The 3D object measurement and recognition system according to claim 3, wherein: The upper platform has at least one X-axis slide rail, one Y-axis slide rail and one Z-axis slide rail, and the image capture units are at least arranged on the X-axis slide rail, or one of the Y-axis slide rail and the Z-axis slide rail.

5. The 3D object measurement and recognition system according to claim 1, wherein: A plurality of positioning devices are arranged on the transparent substrate.

6. The 3D object measurement and recognition system according to claim 5, wherein: The positioning devices are blocking components arranged on the transparent substrate.

7. The 3D object measurement and recognition system according to claim 1, wherein: The image capture unit includes a scanning device and an image capture device, and the image capture device is one of a high-resolution camera or a video camera.

8. The 3D object measurement and recognition system according to claim 1, wherein: The image processing unit includes an image recognition system, and the image recognition system can perform measurement and / or recognition analysis on the captured image.

9. The 3D object measurement and recognition system according to claim 1, wherein: The image processing unit has built-in image processing software and standard information of building materials of doors and windows.

10. The 3D object measurement and recognition system according to claim 9, wherein: The image processing software analyzes the 3D object image using an image comparison module to identify the area in the 3D object image that requires measurement. An edge detection module in the image processing software then obtains the pixel length and width of the area in the 3D object image that requires measurement. This length and width are then compared with pre-stored standard length and width dimensions of building materials for the doors and windows, and then calculated to obtain the actual length and width of the 3D object area in the 3D object image.

11. The 3D object measurement and recognition system according to claim 9, wherein: The image processing software compares the standard information of the building materials of the doors and windows with the 3D object image through an analysis module to identify one of the color and appearance of the 3D object in the image.

12. The 3D object measurement and recognition system according to claim 9, wherein: The image processing software uses an analysis module to compare the standard information of the building materials of the doors and windows with the 3D object image to search for the same accessories at the same position in the 3D object image, and then analyze whether the position of the accessories is correct.

13. The 3D object measurement and recognition system according to claim 2, wherein: At least one light source is provided on one of the upper platform or the lower platform.

Images