CN216132954U

CN216132954U - Product surface defect detection device based on deep learning

Info

Publication number: CN216132954U
Application number: CN202122769190.1U
Authority: CN
Inventors: 黄彭奇子; 段晓君; 黄文伟; 晏良; 陈璇; 肖意可
Original assignee: National University of Defense Technology
Current assignee: National University of Defense Technology
Priority date: 2021-11-12
Filing date: 2021-11-12
Publication date: 2022-03-25
Anticipated expiration: 2031-11-12

Abstract

The utility model provides a product surface defect detection device based on deep learning, which comprises: a conveying belt, a supporting rod and a driving rod; the supporting rods are respectively arranged at the front part and the rear part of the left part of the upper side of the conveyer belt, and the supporting rods are connected with the conveyer belt in a bolt fixing mode; the driving rod is arranged above the space between the supporting rods and is connected with the supporting rods through bearings; the driven cylinders are respectively arranged at the front and the rear of the outer side of the driving rod and are connected with the driving rod in a thread fit manner; the limiting frame is arranged on the outer side of the driven cylinder and is in sliding connection with the driven cylinder; the guide rail is arranged below the support rods and connected with the support rods in a bolt fixing mode. Through structural improvement, have the regulation of being convenient for, quick location and use efficient advantage to the effectual problem and the not enough that appear in having solved current device.

Description

Product surface defect detection device based on deep learning

Technical Field

The utility model relates to the technical field of surface defect detection, in particular to a product surface defect detection device based on deep learning.

Background

In the processing process of a workpiece product, due to various reasons such as raw materials, rolling equipment, processes and the like, defects such as scratches, cracks, rolling marks, scratches, pinholes, oxide scales, holes, pits and the like appear on the surface of the workpiece, and the defects not only affect the appearance of the product, but also reduce the corrosion resistance, wear resistance and fatigue strength of the product, so that after the product is processed, the defect detection is usually performed on the surface of the workpiece.

The structure of present common product surface defect detection device is comparatively simple, and the upside fixed mounting of conveyer belt is one or more image acquisition device usually, lacks corresponding adjustment mechanism, leads to the image information acquisition content of device comprehensive not enough, causes the problem of false retrieval easily.

In view of the above, the present invention provides a product surface defect detection device based on deep learning, which is developed to solve the problems and improve the practical value.

SUMMERY OF THE UTILITY MODEL

The utility model aims to provide a product surface defect detection device based on deep learning, and aims to solve the problems and the defects that the conventional product surface defect detection device in the background art is simple in structure, an image acquisition device is usually fixedly installed on the upper side of a conveying belt, and a corresponding adjusting mechanism is lacked, so that the image information acquisition content of the device is not comprehensive enough, and false detection is easily caused.

In order to achieve the above object, the present invention provides a product surface defect detection device based on deep learning, which is achieved by the following specific technical means:

a product surface defect detection device based on deep learning comprises: the device comprises a conveying belt, a supporting rod, a driving rod, a driven cylinder, a limiting frame, a guide rail, a sliding sleeve, a camera, a fixing plate, an adjusting plate and a screw rod; the supporting rods are respectively arranged at the front part and the rear part of the left part of the upper side of the conveyer belt, and the supporting rods are connected with the conveyer belt in a bolt fixing mode; the driving rod is arranged above the space between the supporting rods and is connected with the supporting rods through bearings; the driven cylinders are respectively arranged at the front and the rear of the outer side of the driving rod and are connected with the driving rod in a thread fit manner; the limiting frame is arranged on the outer side of the driven cylinder and is in sliding connection with the driven cylinder; the guide rail is arranged below the support rods and is connected with the support rods in a bolt fixing mode; the sliding sleeves are arranged in the middle and the front and rear sides of the outer side of the guide rail in a circumferential arrangement mode respectively, the sliding sleeve positioned in the middle is connected with the guide rail in a bolt fixing mode, and the sliding sleeves positioned in the front and rear sides are connected with the guide rail in a sliding mode; the camera is arranged on the lower side of the sliding sleeve and is connected with the sliding sleeve in a bolt fixing mode; the fixed plates are respectively arranged at the front part and the rear part of the right part of the upper side of the conveyer belt, and the fixed plates are connected with the conveyer belt in a bolt fixing mode; the adjusting plate is arranged on the inner side of the fixing plate and is connected with the fixing plate through a screw; the screw rod is provided with two places in the outside of regulating plate, and the screw rod is connected through the welding mode with the regulating plate.

As further optimization of the technical scheme, the driving rod of the product surface defect detection device based on deep learning is a bidirectional spiral screw rod.

As a further optimization of the technical scheme, the driven cylinder of the product surface defect detection device based on deep learning is of a circular tubular structure with threads arranged inside, and a rectangular annular groove is arranged in the middle of the outer side of the driven cylinder.

As a further optimization of the technical scheme, the side of the limiting frame of the product surface defect detection device based on deep learning is regarded as a rectangular ring, and the lower end of the limiting frame is hinged with the sliding sleeves on the front side and the rear side of the device respectively.

As a further optimization of the technical scheme, the product surface defect detection device based on deep learning is characterized in that the guide rail is of an arc-shaped round rod-shaped structure, and the sliding sleeve is of an arc-shaped round tubular structure.

As a further optimization of the technical scheme, the fixing plate of the product surface defect detection device based on deep learning is of a rectangular plate-shaped structure, two circular through holes are formed in the fixing plate, and the fixing plate is connected with the screw rod through a nut.

As further optimization of the technical scheme, the adjusting plate of the product surface defect detecting device based on deep learning is of an arc-shaped plate-shaped structure.

Due to the application of the technical scheme, compared with the prior art, the utility model has the following advantages:

1. according to the product surface defect detection device based on deep learning, the bidirectional spiral screw rod-shaped driving rod is arranged above the supporting rod, the driven barrel and the limiting frame which are connected with the lower sliding sleeve are arranged on the outer side of the driving rod, when the driving rod is rotated, the front camera and the rear camera can be controlled to be synchronously adjusted, so that information of the same plane of a product is acquired in a multi-dimensional mode or multiple planes of the same product are synchronously acquired, and the device has the function of facilitating adjustment.

2. According to the product surface defect detection device based on deep learning, the arc-shaped adjusting plate capable of moving is arranged on the right side of the conveying belt, so that products can be moved to the middle position of the conveying belt to be collected and positioned, and the device has the function of quick positioning.

3. According to the product surface defect detection device based on deep learning, disclosed by the utility model, defect identification is carried out by utilizing a deep learning mode, so that the labor amount of operators can be reduced, the working efficiency can be improved, and the device has an efficient use effect.

4. The device has the advantages of convenient adjustment, quick positioning and high use efficiency through improving the structure of the device, thereby effectively solving the problems and the defects of the prior device.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the utility model and, together with the description, serve to explain the utility model and not to limit the utility model. In the drawings:

FIG. 1 is a schematic structural view of the present invention;

FIG. 2 is a schematic structural view of a driven cylinder according to the present invention;

FIG. 3 is a schematic view of the structure of the sliding sleeve of the present invention;

FIG. 4 is a schematic view of an exploded structure of the adjustment plate according to the present invention;

fig. 5 is a schematic diagram of the information processing of the present invention.

In the figure: conveyer belt 1, bracing piece 2, actuating lever 3, driven section of thick bamboo 4, spacing 5, guide rail 6, sliding sleeve 7, camera 8, fixed plate 9, regulating plate 10, screw rod 11.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.

It is to be noted that, in the description of the present invention, "a plurality" means two or more unless otherwise specified; the terms "upper", "lower", "left", "right", "inner", "outer", "front", "rear", "head", "tail", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, are only for convenience in describing and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, should not be construed as limiting the utility model.

Furthermore, the terms "first," "second," "third," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

Meanwhile, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "connected" and "connected" should be interpreted broadly, for example, as being fixedly connected, detachably connected, or integrally connected; the connection can be mechanical connection or electrical connection; may be directly connected or indirectly connected through an intermediate. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.

Referring to fig. 1 to 5, the present invention provides a specific technical implementation of a product surface defect detection apparatus based on deep learning:

a product surface defect detection device based on deep learning comprises: the device comprises a conveying belt 1, a supporting rod 2, a driving rod 3, a driven cylinder 4, a limiting frame 5, a guide rail 6, a sliding sleeve 7, a camera 8, a fixing plate 9, an adjusting plate 10 and a screw rod 11; the support rods 2 are respectively arranged at the front part and the rear part of the left part of the upper side of the conveyer belt 1, and the support rods 2 are connected with the conveyer belt 1 in a bolt fixing mode; the driving rod 3 is arranged above the supporting rods 2, and the driving rod 3 is connected with the supporting rods 2 through bearings; the driven cylinders 4 are respectively arranged at the front and the rear of the outer side of the driving rod 3, and the driven cylinders 4 are connected with the driving rod 3 in a thread fit mode; the limiting frame 5 is arranged on the outer side of the driven cylinder 4, and the limiting frame 5 is connected with the driven cylinder 4 in a sliding manner; the guide rail 6 is arranged below the support rods 2, and the guide rail 6 is connected with the support rods 2 in a bolt fixing mode; the sliding sleeves 7 are respectively arranged in the middle and the front and rear sides of the outer side of the guide rail 6 in a circumferential arrangement mode, the sliding sleeve 7 positioned in the middle is connected with the guide rail 6 in a bolt fixing mode, and the sliding sleeves 7 in the front and rear sides are connected with the guide rail 6 in a sliding mode; the camera 8 is arranged on the lower side of the sliding sleeve 7, and the camera 8 is connected with the sliding sleeve 7 in a bolt fixing mode; the fixed plates 9 are respectively arranged at the front and the rear of the right part of the upper side of the conveyer belt 1, and the fixed plates 9 are connected with the conveyer belt 1 in a bolt fixing mode; the adjusting plate 10 is arranged on the inner side of the fixing plate 9, and the adjusting plate 10 is connected with the fixing plate 9 through a screw rod 11; two positions of the screw 11 are arranged on the outer side of the adjusting plate 10, and the screw 11 is connected with the adjusting plate 10 in a welding mode.

Specifically, the driving rod 3 is a bidirectional screw rod, the driven cylinder 4 is a circular tubular structure with threads inside, a rectangular ring-shaped groove is arranged in the middle of the outer side of the driven cylinder 4, the side of the limiting frame 5 is regarded as a rectangular ring-shaped structure, the lower end of the limiting frame 5 is respectively hinged with sliding sleeves 7 at the front side and the rear side of the device, the guide rail 6 is an arc-shaped circular rod-shaped structure, and the sliding sleeve 7 is an arc-shaped circular tubular structure, as shown in fig. 1-3, when the driving rod 3 is rotated, the driven cylinder 4 at the outer side of the driving rod is firstly driven to move relatively or oppositely, so as to drive the limiting frame 5 to move synchronously, as the limiting frame 5 is in a rectangular ring shape in side view, and the sliding sleeve 7 is arranged at the outer side of the arc-shaped guide rail 6, therefore, the sliding sleeve 7 hinged with the limiting frame 5 can move synchronously along an arc-shaped path, so as to move the camera 8 synchronously, so as to acquire information of the same plane of a product in multiple dimensions or acquire multiple surfaces of the same product synchronously, the collection effect of the device is improved.

Specifically, fixed plate 9 is rectangular platelike structure, and the inside of fixed plate 9 is equipped with the circular shape through-hole in two places, and fixed plate 9 is connected through the nut with screw rod 11, and regulating plate 10 is curved platelike structure, as shown in fig. 1 and fig. 4, the position of regulating plate 10 can be controlled to the mode that utilizes the nut in the 11 outsides of regulating screw rod, thereby makes the product of conveyer belt 1 upside slide to the middle part of conveyer belt 1 along curved regulating plate 10, conveniently carries out automatic positioning to the product.

Specifically, as shown in fig. 5, after the camera 8 finishes collecting the surface information of the product, the defect conclusion can be automatically obtained and fed back to the operator through the calculation of the defect size and the comparison with the deep learning model, so that the detection efficiency is effectively improved, and the labor cost is saved.

The method comprises the following specific implementation steps:

place the device at surface defect detection process department at first, and insert outside information processing system respectively with camera 8, insert outside drive arrangement with conveyer belt 1, then utilize the mode of adjusting screw 11 outside nut to remove adjusting plate 10's position, make the distance of two department adjusting plate 10 equal with the width of product, finally rotate actuating lever 3, utilize its to driven cylinder 4, spacing 5 and sliding sleeve 7's drive, drive camera 8 and remove to suitable information acquisition position, can follow the right-hand left side removal of conveyer belt 1 with the product and carry out surface defect and detect.

In summary, the following steps: according to the product surface defect detection device based on deep learning, the bidirectional spiral screw rod-shaped driving rod is arranged above the supporting rod, the driven cylinder and the limiting frame which are connected with the lower sliding sleeve are arranged on the outer side of the bidirectional spiral screw rod-shaped driving rod, when the driving rod is rotated, the front camera and the rear camera can be controlled to be synchronously adjusted, so that information of the same plane of a product is acquired in a multi-dimensional mode or multiple surfaces of the same product are synchronously acquired, and the device has the function of being convenient to adjust; the movable arc-shaped adjusting plate is arranged on the right side of the conveying belt, so that products can be moved to the middle position of the conveying belt to be collected and positioned, and the device can be quickly positioned; carry out defect identification through the mode that utilizes degree of depth study, not only can reduce operating personnel's the amount of labour, can also promote work efficiency to make the device play and use the efficient effect, effectively solved the problem that appears in the current device and not enough.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the utility model, the scope of which is defined in the appended claims and their equivalents.

Claims

1. A product surface defect detection device based on deep learning comprises: the device comprises a conveying belt (1), a supporting rod (2), a driving rod (3), a driven cylinder (4), a limiting frame (5), a guide rail (6), a sliding sleeve (7), a camera (8), a fixing plate (9), an adjusting plate (10) and a screw (11); the method is characterized in that: the supporting rods (2) are respectively arranged at the front part and the rear part of the left part of the upper side of the conveyer belt (1), and the supporting rods (2) are connected with the conveyer belt (1) in a bolt fixing mode; the driving rod (3) is arranged above the supporting rods (2), and the driving rod (3) is connected with the supporting rods (2) through bearings; the driven cylinders (4) are respectively arranged at the front and the rear of the outer side of the driving rod (3), and the driven cylinders (4) are connected with the driving rod (3) in a thread fit manner; the limiting frame (5) is arranged on the outer side of the driven cylinder (4), and the limiting frame (5) is connected with the driven cylinder (4) in a sliding mode; the guide rail (6) is arranged below the support rods (2), and the guide rail (6) is connected with the support rods (2) in a bolt fixing mode; the sliding sleeves (7) are arranged in the middle and the front and rear sides of the outer side of the guide rail (6) in a circumferential arrangement mode, the sliding sleeve (7) positioned in the middle is connected with the guide rail (6) in a bolt fixing mode, and the sliding sleeves (7) in the front and rear sides are connected with the guide rail (6) in a sliding mode; the camera (8) is arranged on the lower side of the sliding sleeve (7), and the camera (8) is connected with the sliding sleeve (7) in a bolt fixing mode; the fixed plates (9) are respectively arranged at the front part and the rear part of the right part of the upper side of the conveyer belt (1), and the fixed plates (9) are connected with the conveyer belt (1) in a bolt fixing mode; the adjusting plate (10) is arranged on the inner side of the fixing plate (9), and the adjusting plate (10) is connected with the fixing plate (9) through a screw rod (11); the outer side of the adjusting plate (10) is provided with two positions of the screw rod (11), and the screw rod (11) is connected with the adjusting plate (10) in a welding mode.

2. The device for detecting the surface defects of the product based on the deep learning as claimed in claim 1, wherein: the driving rod (3) is a bidirectional spiral screw rod.

3. The device for detecting the surface defects of the product based on the deep learning as claimed in claim 1, wherein: the driven cylinder (4) is of a circular tubular structure with threads inside, and a rectangular annular groove is formed in the middle of the outer side of the driven cylinder (4).

4. The device for detecting the surface defects of the product based on the deep learning as claimed in claim 1, wherein: the side of the limiting frame (5) is regarded as a rectangular ring, and the lower end of the limiting frame (5) is respectively hinged with the sliding sleeves (7) at the front and the rear of the device.

5. The device for detecting the surface defects of the product based on the deep learning as claimed in claim 1, wherein: the guide rail (6) is of an arc-shaped round rod-shaped structure, and the sliding sleeve (7) is of an arc-shaped round tubular structure.

6. The device for detecting the surface defects of the product based on the deep learning as claimed in claim 1, wherein: the fixing plate (9) is of a rectangular plate-shaped structure, two round through holes are formed in the fixing plate (9), and the fixing plate (9) is connected with the screw (11) through nuts.

7. The device for detecting the surface defects of the product based on the deep learning as claimed in claim 1, wherein: the adjusting plate (10) is of an arc-shaped plate-shaped structure.