CN217277919U - Image training detection device based on meta-learning - Google Patents

Image training detection device based on meta-learning Download PDF

Info

Publication number
CN217277919U
CN217277919U CN202123213208.6U CN202123213208U CN217277919U CN 217277919 U CN217277919 U CN 217277919U CN 202123213208 U CN202123213208 U CN 202123213208U CN 217277919 U CN217277919 U CN 217277919U
Authority
CN
China
Prior art keywords
fixed
base
meta
fixedly connected
detection device
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202123213208.6U
Other languages
Chinese (zh)
Inventor
苏胜
叶锋
施予
林必昌
吕志华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fujian Gabenon Information Technology Co ltd
Original Assignee
Fujian Gabenon Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fujian Gabenon Information Technology Co ltd filed Critical Fujian Gabenon Information Technology Co ltd
Priority to CN202123213208.6U priority Critical patent/CN217277919U/en
Application granted granted Critical
Publication of CN217277919U publication Critical patent/CN217277919U/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The utility model discloses an image training detection device based on meta-learning, which comprises a base, wherein one side of the top end of the base is fixedly provided with a connecting plate, the bottom of one side of the connecting plate is fixedly provided with a detection box, the bottom of the detection box is fixedly provided with a scanning probe, the middle part of the bottom end of the base is fixedly provided with a cylinder, the output end of the cylinder passes through the fixedly connected fixed block of the base, the two sides of the fixed block are both rotatably provided with connecting rods, and the ends of the two connecting rods departing from the fixed rod are both rotatably connected with a special-shaped driving rod, the utility model relates to an image training detection device based on meta-learning, through the mutual cooperation of the cylinder, a mounting block, the connecting rods and the special-shaped driving rod, a slide block is driven to move towards the middle part of a strip-shaped hole, namely a fixed disk and a fixed disk are driven to clamp, a driving motor drives the fixed disk to rotate after clamping, namely drives the fixed disk to rotate, and does not need manual turnover of workers, the detection efficiency is improved.

Description

Image training detection device based on meta-learning
Technical Field
The utility model relates to an image detection field specifically is an image training detection device based on meta-learning.
Background
The english name of meta learning is MetaLearning, which is "how to learn the society" compared with the traditional machine learning research mode, namely, the learning of a new task is guided by using the prior knowledge and experience, the learning ability is learned, the goal of meta-learning is to learn a model on various different learning tasks, so that some new learning tasks can be solved by using a small amount of samples, the challenge of this task is that the model needs to combine a small amount of sample information from previous experience and the current new task, and avoid overfitting on the new data, the acquisition of sample image information needs to use an image training detection device, the existing image training detection device only acquires images of workpieces through a computer, in the actual use process, the workpiece image is not completely acquired, the workpiece image needs to be manually turned over, and the acquisition efficiency is poor.
SUMMERY OF THE UTILITY MODEL
An object of the utility model is to provide an image training detection device based on meta-learning to solve the current image training detection device who proposes in the above-mentioned background art and only carry out image acquisition to the work piece through the computer, it is incomplete to work piece image acquisition in the in-service use process, need manual upset to it, problem that detection efficiency is poor.
In order to achieve the above object, the utility model provides a following technical scheme: an image training detection device based on meta-learning comprises a base, a connecting plate is fixedly arranged on one side of the top end of the base, the bottom of one side of the connecting plate is fixedly provided with a detection box, the bottom of the detection box is fixedly provided with a scanning probe, the middle part of the bottom end of the base is fixedly provided with an air cylinder, the output end of the air cylinder penetrates through the base and is fixedly connected with a fixed block, connecting rods are rotatably arranged on both sides of the fixed block, one end of each connecting rod, which is far away from the fixed rod, is rotatably connected with a special-shaped driving rod, the middle parts of the two special-shaped driving rods are fixedly connected with rotating shafts, the two rotating shafts are rotatably connected with the inner wall of the detection box, a strip-shaped hole is formed in the middle of the top end of the base, two sliding blocks are arranged in the strip-shaped hole in a sliding mode, and the bottoms of the two sliding blocks are respectively connected with one ends, deviating from the connecting rod, of the two special-shaped driving rods in a rotating mode.
Preferably, two the top of slider is first fixed plate of fixedly connected with and second fixed plate respectively, one side fixed mounting that first fixed plate deviates from the second fixed plate has driving motor, first fixed plate fixedly connected with fixed disk is passed to driving motor's output, drives first fixed disk through driving motor and rotates.
Preferably, one side fixedly connected with bearing that second fixed plate and first fixed plate are relative, one side that the bearing deviates from first fixed plate rotates and is connected with fixed disk two, and the fixed disk presss from both sides tightly with a pair of work piece of fixed disk, when fixed disk one rotates, can drive fixed plate dish two and rotate thereupon.
Preferably, the rubber cushions are fixedly connected to one opposite sides of the first fixed disk and the second fixed disk, and the first fixed disk and the second fixed disk are prevented from damaging a workpiece due to overlarge clamping force in the workpiece clamping process.
Preferably, the display screen is embedded in the middle of the front face of the detection box, a plurality of control buttons are fixedly arranged on one side of the front face of the detection box, and collected image information can be conveniently checked through the display screen.
Preferably, four equal fixedly connected with support columns in corner of base bottom, four the equal fixedly connected with fixed bolster in bottom of support column supports whole device through the support column and makes things convenient for carrying wholly.
Compared with the prior art, the beneficial effects of the utility model are that: put the work piece between fixed disk one and fixed disk two, through the cylinder, the installation piece, mutually supporting of connecting rod and dysmorphism actuating lever, the cylinder drives the fixed block rebound, the fixed block drives the one end of two connecting rods upwards, the other end of connecting rod drives the one end angle modulation downwards of dysmorphism actuating lever, the other end of dysmorphism actuating lever rotates round the pivot, thereby it removes to bar hole middle part to drive the slider, first fixed plate and second fixed plate drive fixed disk one and fixed disk two are close to each other and press from both sides the work piece promptly, press from both sides tight back start driving motor, driving motor drives fixed plate one and rotates, it rotates promptly to drive the work piece, can the multi-angle carry out image acquisition to the work piece and detect, whole degree of mechanization and detection efficiency are high, do not need the manual upset of staff.
Drawings
Fig. 1 is a perspective view of the present invention;
fig. 2 is a cross-sectional view of the base of the present invention;
fig. 3 is a partial connection diagram of the present invention.
In the figure: 1. a base; 2. a connecting plate; 3. a detection box; 4. scanning the probe; 5. a cylinder; 6. a fixed block; 7. a connecting rod; 8. a special-shaped driving rod; 9. a rotating shaft; 10. a strip-shaped hole; 11. a slider; 12. a first fixing plate; 13. a second fixing plate; 14. a drive motor; 15. a first fixed disc; 16. a bearing; 17. fixing a second disc; 18. a rubber cushion; 19. a display screen; 20. a control button; 21. and (7) supporting the column.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
Please refer to fig. 1-3, the utility model provides an image training detection device based on first study, including base 1, one side fixed mounting on base 1 top has connecting plate 2, the bottom fixed mounting of 2 one sides of connecting plate has detection case 3, the fixed scanning probe 4 that is provided with in bottom of detection case 3, the middle part fixed mounting of base 1 bottom has cylinder 5, the output of cylinder 5 passes base 1 fixedly connected with fixed block 6, the both sides of fixed block 6 are all rotated and are provided with connecting rod 7, two connecting rod 7 deviate from the one end of dead lever and all rotate and be connected with special-shaped actuating lever 8, the equal fixedly connected with pivot 9 in middle part of two special-shaped actuating lever 8, two pivot 9 all are connected with the inner wall rotation of detection case 3, bar hole 10 has been seted up at the middle part on base 1 top, the inside slip of bar hole 10 is provided with two sliders 11, the bottom of two sliders 11 rotates with the one end that two special-shaped actuating lever 8 deviate from 7 respectively and is connected.
During the use, put the work piece between fixed disk one 15 and fixed disk two 17, start cylinder 5, the output of cylinder 5 drives fixed block 6 and rises, the rising of fixed block 6 drives the one end of connecting rod 7 and rises, the other end of connecting rod 7 drives the bottom of dysmorphism actuating lever 8 and rotates around middle part pivot 9, meanwhile, the top of dysmorphism actuating lever 8 drives slider 11 and slides in bar hole 10 along with its bottom rotation, it is close to the work piece gradually to drive first fixed plate 12 and second fixed plate 13 in the slip process, until closely laminating with the work piece, can carry out the plane of sweeping to the work piece through scanning probe 4, scan image information on the work piece, train the detection through detection case 3 to the information that scans.
The top of two sliders 11 is first fixed plate 12 and second fixed plate 13 of fixedly connected with respectively, one side fixed mounting that first fixed plate 12 deviates from second fixed plate 13 has driving motor 14, first fixed plate 12 fixedly connected with fixed disk 15 is passed to driving motor 14's output, second fixed plate 13 and one side fixedly connected with bearing 16 that first fixed plate 12 is relative, bearing 16 deviates from one side rotation of first fixed plate 12 and is connected with fixed disk second 17, the equal fixedly connected with rubber cushion 18 in one side that fixed disk first 15 and fixed disk second 17 are relative.
In the scanning process, the driving motor 14 drives the first fixed disk 15 to rotate, and the first fixed disk 15 and the second fixed disk 17 are in a clamping state with the workpiece, so that when the first fixed disk 15 rotates, the workpiece and the second fixed disk 17 can be driven to synchronously rotate, mechanical overturning can be realized, the scanning probe 4 can synchronously acquire image information at the bottom of the workpiece, manual overturning is not needed, the integral mechanization degree and the working efficiency are improved, and the labor cost is reduced.
The middle part of the front of the detection box 3 is embedded with a display screen 19, and one side of the front of the detection box 3 is fixedly provided with a plurality of control buttons 20.
The acquired image information can be directly visualized through the display screen 19, and the control buttons 20 can control the detection box 3.
The equal fixedly connected with support column 21 in four corners of base 1 bottom, the equal fixedly connected with fixed bolster in bottom of four support columns 21.
The whole base 1 with the whole device can be supported by the supporting columns 21, so that the base can be conveniently carried in the using process.
When the embodiment of the application is used: a workpiece is placed between a first fixed disk 15 and a second fixed disk 17, an air cylinder 5 is started, the output end of the air cylinder 5 drives a fixed block 6 to ascend, the ascending of the fixed block 6 drives one end of a connecting rod 7 to ascend, the other end of the connecting rod 7 drives the bottom end of a special-shaped driving rod 8 to rotate around a middle rotating shaft 9, meanwhile, the top end of the special-shaped driving rod 8 drives a sliding block 11 to slide in a strip-shaped hole 10 along with the rotation of the bottom end of the special-shaped driving rod, a first fixed plate 12 and a second fixed plate 13 are driven to gradually approach the workpiece in the sliding process until the special-shaped driving rod is tightly attached to the workpiece, the workpiece can be scanned through a scanning probe 4, image information on the workpiece is scanned, the scanned information is trained and detected through a detection box 3, in the scanning process, the driving motor 14 drives the first fixed disk 15 to rotate, because the first fixed disk 15 and the second fixed disk 17 are in a clamping state with the workpiece, the workpiece and the second fixed disk 17 can be driven to synchronously rotate, and mechanical turnover can be realized, so that the scanning probe 4 can synchronously acquire image information at the bottom of the workpiece without manually overturning the workpiece, the overall degree of mechanization and the working efficiency are improved, and the labor cost is reduced.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments or portions thereof without departing from the spirit and scope of the invention.

Claims (6)

1. The utility model provides an image training detection device based on meta-learning, includes base (1), its characterized in that: the detection device is characterized in that a connecting plate (2) is fixedly mounted on one side of the top end of the base (1), a detection box (3) is fixedly mounted on the bottom of one side of the connecting plate (2), a scanning probe (4) is fixedly arranged on the bottom of the detection box (3), an air cylinder (5) is fixedly mounted in the middle of the bottom end of the base (1), the output end of the air cylinder (5) penetrates through the base (1) and is fixedly connected with a fixed block (6), connecting rods (7) are rotatably arranged on two sides of the fixed block (6), one ends of the two connecting rods (7) departing from the fixed rod are rotatably connected with special-shaped driving rods (8), rotating shafts (9) are fixedly connected in the middle of the two special-shaped driving rods (8), the two rotating shafts (9) are rotatably connected with the inner wall of the detection box (3), a strip-shaped hole (10) is formed in the middle of the top end of the base (1), and two sliding blocks (11) are slidably arranged inside the strip-shaped hole (10), the bottoms of the two sliding blocks (11) are respectively connected with one ends of the two special-shaped driving rods (8) departing from the connecting rod (7) in a rotating mode.
2. The meta-learning based image training detection device of claim 1, wherein: two the top of slider (11) is fixedly connected with first fixed plate (12) and second fixed plate (13) respectively, one side fixed mounting that first fixed plate (12) deviates from second fixed plate (13) has driving motor (14), first fixed plate (12) fixedly connected with fixed disk (15) is passed to the output of driving motor (14).
3. The meta-learning based image training detection apparatus according to claim 2, wherein: second fixed plate (13) and the relative one side fixedly connected with bearing (16) of first fixed plate (12), one side that bearing (16) deviates from first fixed plate (12) is rotated and is connected with fixed disk two (17).
4. The meta-learning based image training detection device of claim 2, wherein: and one opposite sides of the first fixed disk (15) and the second fixed disk (17) are fixedly connected with rubber cushions (18).
5. The meta-learning based image training detection device of claim 1, wherein: the front middle part of the detection box (3) is embedded with a display screen (19), and one side of the front side of the detection box (3) is fixedly provided with a plurality of control buttons (20).
6. The meta-learning based image training detection device of claim 1, wherein: the base is characterized in that four corners of the bottom end of the base (1) are fixedly connected with supporting columns (21), and four bottoms of the supporting columns (21) are fixedly connected with fixing pads.
CN202123213208.6U 2021-12-20 2021-12-20 Image training detection device based on meta-learning Active CN217277919U (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202123213208.6U CN217277919U (en) 2021-12-20 2021-12-20 Image training detection device based on meta-learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202123213208.6U CN217277919U (en) 2021-12-20 2021-12-20 Image training detection device based on meta-learning

Publications (1)

Publication Number Publication Date
CN217277919U true CN217277919U (en) 2022-08-23

Family

ID=82889444

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202123213208.6U Active CN217277919U (en) 2021-12-20 2021-12-20 Image training detection device based on meta-learning

Country Status (1)

Country Link
CN (1) CN217277919U (en)

Similar Documents

Publication Publication Date Title
CN204453789U (en) A kind of automatic labeling and granting vacuum test tube device
EP3392660A1 (en) Array-type feces occult blood detection analyzer
CN102735855A (en) Fully automatic leucorrhea inspection analyzer
CN217277919U (en) Image training detection device based on meta-learning
CN207456685U (en) One kind is used for intellective dust collector handle test equipment
CN202676714U (en) Full-automatic leucorrhea examination analyzer
CN206725468U (en) A kind of glass defect detection equipment
CN209673818U (en) A kind of detection scanning means
CN213517223U (en) Jig for detecting circuit board bonding pad
CN106370876B (en) A kind of indirect immunofluorescence automatic business processing instrument
CN210833961U (en) Spring elasticity detection device
CN210741341U (en) Product inner hole diameter detection equipment
CN106443036A (en) Sample rack bearing device provided with double-side opening and closing mechanism
CN209673541U (en) A kind of leather quality detection device
CN207487569U (en) A kind of motor testing apparatus for verticality
CN206531860U (en) A kind of multi-joint card gold mark reading module
CN112675778A (en) Batching device of blood testing machine
CN219284780U (en) Test piece processing mechanism of full-automatic pressure testing machine
CN215491685U (en) Intelligent detection device based on image analysis
CN217505164U (en) Efficient organism testing arrangement
CN211710383U (en) Glass coding machine
CN218298059U (en) Module appearance defect detection device capable of automatically adjusting light source
CN213593994U (en) Display device for analyzing and processing bridge monitoring data
CN213844464U (en) Display device for teaching of amplified signal diagnosis experiment material
CN209765203U (en) Multifunctional detection jig for liquid crystal LED display screen

Legal Events

Date Code Title Description
GR01 Patent grant
GR01 Patent grant