Standing type engineering apparent quality detector based on binocular vision

Abstract

The utility model relates to a binocular vision-based stand type engineering apparent quality detector, which comprises a support frame and a horizontal slide rail, the middle part of the lower side of the horizontal slide rail is connected to the top of the support frame and can horizontally rotate around the support frame, and the middle part of the upper side of the horizontal slide rail is connected with a controller; the horizontal sliding rail is connected with a first adjusting sliding block and a second adjusting sliding block which are located on the two sides of the controller and can slide relative to the horizontal sliding rail, the first adjusting sliding block is provided with a first camera, the second adjusting sliding block is provided with a second camera, and the first camera and the second camera are RGB binocular depth cameras. The erecting and standing type engineering apparent quality detector based on binocular vision is reasonable in structural design, convenient and practical, has high-efficiency localization processing capability, realizes variable visual field of the camera, and meets the detection requirements of large and small scenes.

Description

TECHNICAL FIELD

[0001] The utility model relates to engineering detection technical field especially a kind of stand type engineering apparent quality detector based on binocular vision. BACKGROUND

[0002] With the rapid development of computer vision, artificial intelligence and other technologies, this way of obtaining the position information of target objects in space through depth cameras has been widely used in automatic driving, three-dimensional reconstruction, industrial defect detection, robot navigation and other fields. At the same time, intelligent detection technology based on machine vision has also begun to be applied in the field of civil engineering, especially in the detection of engineering apparent quality.

[0003] Depth cameras can be roughly divided into the following three categories according to their working principles: TOF, 3D structured light and RGB binoculars.

[0004] TOF (Time-Of-Flight) depth cameras obtain scene depth information by measuring the flight time of light between the target and the camera. Patent 202410536064.7 fuses the image information collected by TOF cameras and RGB cameras to carry out distance measurement work, and simultaneously splices multiple groups of fused images and compares them with tunnel construction technical drawings; Patent 202311737586.5 performs segmentation operation on the three-dimensional point cloud collected by TOF cameras with the help of semantic segmentation, estimates the normal vector, converts it into the normal vector under the base coordinate system of the mechanical arm through coordinate system conversion, and then completes the normal alignment of the building construction robot working surface; Patent 202211331064.0 performs cropping and point cloud registration on the truck empty point cloud and loaded ore rock loading point cloud obtained by TOF cameras through image segmentation, so that the length-width plane information corresponds, and uses stereo geometry algorithm to calculate the ore rock loading volume.

[0005] 3D structured light depth camera projects a specific structured light pattern (stripes, dot matrix, etc.) onto the target object, captures the deformed pattern reflected by the object surface with a camera, and calculates the object depth information according to the changes in the reflected pattern. Patent 202410531549.7 uses least squares method to fit the spatial point cloud data obtained by structured light camera to an elliptic cylindrical surface, and takes the fitted surface as the reference surface for tunnel flatness detection by flatness algorithm; Patent 202311355814.2 uses structured light camera to scan the shield tunnel segment, and extracts, learns and trains the crack features for high-resolution images, and enhances the image for low-resolution images, thereby reducing the requirement for image resolution, and realizing intelligent identification of shield tunnel segment cracks; Patent 202211483582.4 eliminates the influence of vehicle vibration on the road surface image collected by 3D structured light camera, and fuses with the road surface image, and completes the three-dimensional reconstruction and volume calculation of road rut disease by constructing three-dimensional space matrix and plane fault cutting.

[0006] RGB binocular depth camera calculates the object depth information by obtaining the parallax (i.e. the difference in position of the same object in two images) between the images taken by the two camera lenses, and using the principle of triangulation. Patent 202410561239.X enables the double compound eye to obtain sub-eye images of the same target in the object space through optical axis correction, feature point search and matching, and data processing, thereby realizing simultaneous ranging of multiple targets in a wide range of scenes; Patent 202311833517.4 and Patent 202311828749.0 calculate and judge the position of feature points in the image, adjust the camera pose according to the difference between the feature points and the target set position, and then accurately calculate the roadway deformation, realizing roadway deformation monitoring; Patent 202310832672.8 calibrates the camera and extracts the feature points of the decoded image, and designs a binocular stereo vision three-dimensional reconstruction system based on wavefront coding through steps such as stereo matching, depth calculation and three-dimensional space point reconstruction.

[0007] In terms of ranging method, RGB binocular depth camera is passive, and compared with TOF and 3D structured light depth camera, it does not need transmitter and receiver, so the hardware device is relatively cheap; in terms of measurement accuracy, RGB binocular depth camera can achieve millimeter-level measurement accuracy in close range, which can meet the engineering measurement demand. However, RGB binocular camera usually does not have local processing capability, and the patents mentioned above usually need to rely on external computer for data processing and analysis, and the real-time processing efficiency is poor, which limits its application in civil engineering field to some extent. The portable binocular camera involved in Patent 202420197345.X can process and display the image in time, but the binocular camera module is limited by the size of the camera, and is not suitable for engineering large scene measurement. Utility model content

[0008] Therefore, the utility model discloses a structure design is reasonable, and the stand type engineering apparent quality detector based on binocular vision has convenient and practical, possesses the localization processing ability, realizes camera field of vision variable.

[0009] The utility model discloses the following scheme realizes: a stand type engineering apparent quality detector based on binocular vision, including support frame and horizontal slide rail, horizontal slide rail lower side middle part is connected in support frame top and can rotate along horizontal around support frame, horizontal slide rail upper side middle part is connected with controller, and horizontal slide rail is connected with the first adjusting sliding block and second adjusting sliding block located controller both sides and can slide relative to horizontal slide rail, be equipped with the first camera on the first adjusting sliding block, be equipped with the second camera on the second adjusting sliding block, first camera and second camera are RGB binocular depth camera.

[0010] Further, the first camera and the second camera are respectively connected with the first adjusting sliding block and the second adjusting sliding block through a spherical hinge.

[0011] Further, the support frame includes a fixed sleeve, three telescopic supporting legs are connected to the lower end of the fixed sleeve; a rotating shaft is fixedly connected to the lower middle part of the horizontal slide rail and extends into the fixed sleeve and can rotate relative to the fixed sleeve, and a locking bolt A is connected through one side of the fixed sleeve to abut against the rotating shaft and limit the rotation of the rotating shaft.

[0012] Further, the first adjusting sliding block and the second adjusting sliding block are respectively connected through a locking bolt B to abut against the horizontal slide rail and limit the sliding of the first adjusting sliding block and the second adjusting sliding block.

[0013] Further, the supporting leg includes an outer tube and an inner tube which are telescopically connected and can slide relative to each other, the inner tube extends from the lower end of the outer tube, a limiting hole is formed in the lower end of the outer tube, a latch is arranged in the limiting hole, and a plurality of pin holes are distributed on the inner tube and selectively cooperate with the latch.

[0014] Further, the controller includes a casing composed of an upper shell and a lower shell, a power supply and a computer mainboard are arranged in the casing, a USB interface and an Ethernet interface are arranged on the computer mainboard and exposed from the side of the casing, and a display screen is arranged on the top of the casing.

[0015] Compared with the prior art, the utility model has the following beneficial effects: the stand type engineering apparent quality detector based on binocular vision of the utility model has reasonable structure design, is convenient and practical, has efficient localization processing capability, realizes variable camera field of vision, and meets the detection requirements of large and small scenes.

[0016] In order to make the purpose, technical scheme and advantages of the utility model more clear and obvious, the following will be further described in detail through specific embodiments and related drawings. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall structure of an embodiment of this utility model;

[0018] Figure 2 This is a partial structural schematic diagram of an embodiment of the present utility model;

[0019] Figure 3 This is an exploded view of the controller according to an embodiment of this utility model;

[0020] Figure 4 This is a schematic diagram of the connection between the horizontal slide rail and the support frame in an embodiment of this utility model;

[0021] Figure 5 This is a partial sectional view of the support leg of the support frame according to an embodiment of the present utility model;

[0022] The following are the labels in the diagram: 100-Support frame, 110-Fixing sleeve, 111-Locking bolt A, 120-Support foot, 121-Outer tube, 122-Inner tube, 123-Pin, 200-Horizontal slide rail, 210-Rotating shaft, 300-Controller, 310-Upper housing, 320-Lower housing, 330-Power supply, 340-Computer motherboard, 350-USB interface, 360-Ethernet interface, 370-Display screen, 400-First adjusting slider, 410-Locking bolt B, 500-Second adjusting slider, 600-First camera, 700-Second camera, 800-Spherical hinge. Detailed Implementation

[0023] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0024] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0025] like Figures 1-5As shown, a kind of engineering apparent quality detector of stand type based on binocular vision, including support frame 100 and horizontal slide rail 200, horizontal slide rail 200 lower middle part is connected to the top of support frame and can be rotated along horizontal around support frame, the upper middle part of the horizontal slide rail is connected with controller 300, horizontal slide rail is connected with the first adjusting slider 400 and second adjusting slider 500 located at the two sides of controller and can slide relative to horizontal slide rail, first camera 600 is equipped on the first adjusting slider, second camera 700 is equipped on the second adjusting slider, the first camera and second camera are RGB binocular depth camera.The utility model engineering apparent quality detector of stand type by controller 300 as core driving element, is responsible for controlling first camera and second camera gather engineering apparent quality image, and carries out relevant image processing.Combined with the advantages of RGB binocular depth camera and computer mainboard, utilize the principle of RGB binocular depth camera parallax to realize engineering apparent quality accurate detection, avoid using additional sensing equipment, and the hardware cost is lower.At the same time, by controller integrated control, without relying on external computer processing data, realize efficient localization processing capacity;First adjusting slider and second adjusting slider are assembled on horizontal slide rail and move horizontally, so that the field of view of first camera and second camera is variable;Horizontal slide rail can rotate around support frame, and it is convenient for first camera and second camera to shoot the scene around, realize that the field of view of camera is variable, and the detection needs of big and small scenes are considered.

[0026] In the embodiment, the first camera 600 and the second camera 700 are connected to the first adjusting slider 400 and the second adjusting slider 500 respectively through a ball hinge. The camera and the corresponding adjusting slider are connected through the ball hinge to ensure that the camera can rotate around the ball hinge for self-angle adjustment. The first camera and the second camera have a common field of view when collecting images through the hinged adjustment.

[0027] In the embodiment, the support frame 100 includes a fixed sleeve 110, and the lower end of the fixed sleeve is connected with three telescopic supporting legs 120. The lower middle part of the horizontal slide rail 200 is fixedly connected with a rotating shaft 210 that extends into the fixed sleeve and can rotate relative to the fixed sleeve. The fixed sleeve 110 is connected with a locking bolt A111 on one side to resist the rotating shaft and limit the rotation of the rotating shaft.

[0028] In the embodiment, the first adjusting slider 400 and the second adjusting slider 500 are respectively connected with a locking bolt B410 that resists the horizontal slide rail and limits the sliding of the first adjusting slider and the second adjusting slider. The first adjusting slider 400 and the second adjusting slider 500 are 70 mm long, 50 mm wide and 15 mm high. The horizontal slide rail can provide a sliding length of 700 mm for the first adjusting slider 400 and the second adjusting slider 500.

[0029] In the embodiment, the supporting leg 120 comprises two sections of outer tube 121 and inner tube 122 which are connected and can slide relative to each other, the inner tube extends from the lower end of the outer tube, the lower end of the outer tube is provided with a limiting hole, the limiting hole is provided with a bolt 123, and the inner tube is provided with pin holes which are matched with the bolt.

[0030] In the embodiment, the controller comprises a casing which is composed of an upper shell 310 and a lower shell 320, the upper shell 310 is provided with a plurality of protruding buckles at the edge thereof, and the lower shell 320 is provided with matching buckle slots at the corresponding positions thereof. The casing is provided with a power supply 330 and a computer mainboard 340, the computer mainboard is provided with a USB interface 350 and an Ethernet interface 360 which are exposed from the side of the casing, the upper shell 310 and the lower shell 320 form an internal structure with a certain accommodating space therebetween, the power supply stably supplies power to the computer mainboard to meet the long-time use requirement, the computer mainboard is provided with the USB interface and the Ethernet interface to meet the detection instrument debugging and external data interaction requirement, the first camera and the second camera are connected with the USB interface of the computer mainboard through an external USB data line to realize image data transmission.

[0031] The size of the controller is: length 200 mm, width 150 mm, height 65 mm; the size of the display screen is: length 180 mm, width 100 mm, height 4 mm; the size of the computer mainboard is: length 180 mm, width 100 mm, height 50 mm; the size of the power supply 505 is: length 160 mm, width 40 mm, height 60 mm.

[0032] In the computer mainboard operating system, camera driver program, image processing algorithm and interface display program are burned to realize the collection, processing and result display of the engineering appearance quality image. After the image collection of the first camera and the second camera is completed, image processing is performed to provide reliable data basis for subsequent engineering geometric information extraction. The image processing algorithm mainly includes feature extraction, region generation, target classification, mask generation and post-processing operations. Among them, the feature extraction generates a feature map containing image semantic information and spatial information to provide a basis for subsequent detection; the region generation filters out the target region; the target classification further determines the target category contained in the region (such as distinguishing steel bars, templates, etc.); the pixel-level mask is further generated for each target category to accurately describe the target object contour and shape; the post-processing further processes the size (such as steel bar diameter, template length) and other parameter information according to the specific detection requirements, and finally transmits the detection result to the display screen for intuitive evaluation by the detection personnel. The computer program burned into the microcomputer mainboard is a known technology in the art, which is realized by using the known technology of those skilled in the art and does not belong to the protection scope of the present utility model, therefore, it will not be further described here. The present utility model protects the structure of the stand-type engineering appearance quality detector, realizes the combination of the RGB binocular depth camera and the computer mainboard, and the camera field of view is variable.

[0033] The top of the shell is provided with a display screen 370 connected with the computer mainboard through an SPI interface, and the final image processing result is displayed on the display screen in real time, so that the detection personnel can intuitively view.

[0034] Unless otherwise stated, if the above-mentioned any technical solution of the present utility model discloses a numerical range, the disclosed numerical range is a preferred numerical range. Any person skilled in the art should understand that the preferred numerical range is only one of the many implementable values with more obvious technical effects or representative values. Since there are too many values, it is impossible to enumerate them all, therefore, the present utility model only discloses part of the values to illustrate the technical solutions of the present utility model, and the above-mentioned enumerated values should not constitute a limitation on the protection scope of the present utility model.

[0035] If the present utility model discloses or involves mutually fixed and connected parts or structural members, unless otherwise stated, the fixed connection can be understood as: detachable fixed connection (for example, connected by using bolts or screws), and can also be understood as: non-detachable fixed connection (for example, riveting, welding), of course, the mutual fixed connection can also be replaced by an integral structure (for example, manufactured by using casting process integral forming) (except for obvious cases that cannot use integral forming process).

[0036] In addition, the terms used to represent the positional relationship or shape in any of the technical solutions of the utility model disclosed above include the states or shapes similar, analogous or close to the terms unless otherwise stated.

[0037] Any component provided by the utility model can be assembled from multiple individual components or manufactured as an individual component by an integral forming process.

[0038] The above is only a preferred embodiment of the utility model, and does not limit the utility model in other forms. Any skilled person in the art can modify or change the above disclosed technical content to obtain equivalent embodiments. However, any simple modification, equivalent change and modification of the above embodiments without departing from the technical solution of the utility model and according to the technical essence of the utility model still belong to the protection scope of the technical solution of the utility model.

Claims

1. A stand-alone engineering appearance quality inspection instrument based on binocular vision, characterized in that: It includes a support frame and a horizontal slide rail. The lower middle part of the horizontal slide rail is connected to the top of the support frame and can rotate horizontally around the support frame. The upper middle part of the horizontal slide rail is connected to a controller. The horizontal slide rail is connected to a first adjustment slider and a second adjustment slider located on both sides of the controller and can slide relative to the horizontal slide rail. The first adjustment slider is equipped with a first camera, and the second adjustment slider is equipped with a second camera. The first camera and the second camera are RGB binocular depth cameras.

2. The binocular vision-based stationary engineering appearance quality inspection instrument according to claim 1, characterized in that: The first camera and the second camera are respectively connected to the first adjusting slider and the second adjusting slider via ball joints.

3. The binocular vision-based stationary engineering appearance quality inspection instrument according to claim 1, characterized in that: The support frame includes a fixed sleeve, the lower end of which is connected to three retractable support feet; a rotating shaft that extends into the fixed sleeve and can rotate relative to it is fixedly connected to the middle of the lower side of the horizontal slide rail, and a locking bolt A is connected through one side of the fixed sleeve to resist the rotating shaft and limit its rotation.

4. The stand-mounted engineering appearance quality inspection instrument based on binocular vision according to claim 1, characterized in that: Locking bolts B are respectively connected through the first and second adjusting sliders to resist the horizontal slide rail and limit the sliding of the first and second adjusting sliders.

5. The stand-mounted engineering appearance quality inspection instrument based on binocular vision according to claim 3, characterized in that: The support leg includes two sleeved outer tubes and inner tubes that can slide relative to each other. The inner tube extends from the lower end of the outer tube. A limiting hole is opened at the lower end of the outer tube. A pin is inserted into the limiting hole. The inner tube has pin holes distributed at intervals that can be matched with the pin.

6. The stand-mounted engineering appearance quality inspection instrument based on binocular vision according to claim 1, characterized in that: The controller includes a housing consisting of an upper shell and a lower shell. The housing contains a power supply and a computer motherboard. The computer motherboard has a USB interface and an Ethernet interface exposed from the side of the housing. The top of the housing has a display screen.

Images