CN112345551B - Method and system for detecting surface defects of inner wall and outer wall of aircraft conduit bell mouth - Google Patents
Method and system for detecting surface defects of inner wall and outer wall of aircraft conduit bell mouth Download PDFInfo
- Publication number
- CN112345551B CN112345551B CN202011225991.5A CN202011225991A CN112345551B CN 112345551 B CN112345551 B CN 112345551B CN 202011225991 A CN202011225991 A CN 202011225991A CN 112345551 B CN112345551 B CN 112345551B
- Authority
- CN
- China
- Prior art keywords
- wall
- light
- defects
- detection
- conical mirror
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/9515—Objects of complex shape, e.g. examined with use of a surface follower device
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Instruments For Viewing The Inside Of Hollow Bodies (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention belongs to the field of surface defect detection of aerospace precise micro parts, and discloses a method and a system for detecting surface defects of the inner wall and the outer wall of a horn mouth of an aircraft guide pipe, wherein an inner conical mirror is used for reflecting light guided by optical fibers to the outer wall of a workpiece uniformly in cooperation with a lamp strip, and the outer wall of the workpiece reflects the light to a lens for imaging through the same light path; the endoscopic imaging head with a light source is adopted to be matched with a reflection system for illumination, so that the large-scale uniform irradiation is realized, and the clear imaging of the curved surfaces of the inner wall and the outer wall is ensured; the endoscopic imaging head is imaged by the objective lens and transmitted to the CCD rake surface, then the CCD converts the optical signal into an electronic signal, the data is transmitted to the video endoscopic control group, and the image is output to the computer by the control group; and (3) training a target detection model by applying a deep learning method, and detecting defects in the image. The detection requirements of the inner wall and the outer wall of the bell mouth of the conduit with the diameter of phi 4-phi 10 can be met, and the device is convenient to use, accurate and rapid in detection; has certain universality.
Description
Technical Field
The invention belongs to the field of surface defect detection of aerospace precise micro parts, and particularly relates to a method and a system for detecting surface defects of inner and outer walls of a horn mouth of an aircraft guide pipe.
Background
At present, the bell mouth of the pipe of the oil retention machine is used as an important carrier for oil transportation of the oil retention machine, and plays an important role in the oil retention machine. The surface quality detection of the conduit bell mouth can ensure the long-term stable operation of the airplane, and the best performance can be exerted in the aspects of service life, economy, power performance and the like, so that the operation failure of the airplane caused by the unqualified surface quality of the conduit bell mouth is prevented. The production process of the conduit bell mouth is difficult, and under the action of factors such as long-term high temperature, high pressure, fatigue and the like, the surface of the conduit bell mouth has defects such as collision, cracks, scratches and the like, and even endangers the flight safety. The defects of the aircraft oil leakage and insufficient fuel supply are caused, and the unexpected huge loss is caused. Therefore, it is necessary to perform surface defect detection on the aircraft conduit bell.
At present, there are various methods for non-contact detection of inner and outer walls of small-bore workpieces. Mainly, there are industrial endoscopy, eddy current testing, laser speckle method, ultrasonic method, and the like. Other methods include an optical single-point scanning method and a magnetic flux leakage method.
Industrial endoscopes are a common solution, but their measurement repeatability is difficult to guarantee, is not conducive to preserving images, and is expensive. If the inner wall surface of the workpiece has defects, the appearance of the workpiece is influenced, and the function, the stability, the safety and the like of the workpiece are greatly influenced.
In China, the research of the eddy current flaw detection technology starts from 60 s of the twentieth century, and sets of eddy current flaw detector equipment are developed and manufactured in 80 s, and relevant standards are established. The principle of eddy current detection is that an alternating current electromagnetic coil is used for inducing eddy current on the surface of a metal component, and when the inner wall of a workpiece has a recess, a protrusion or other defects, the magnitude of current change in the coil can be changed, so that the detection purpose is achieved. The eddy current flaw detection method has the advantages of no need of a coupling agent, high detection sensitivity, high detection speed, non-contact property and the like, and the size can be very small. Therefore, the method is widely applied to detecting the mass of different kinds of metal members, such as airplane hubs, steel pipes, wire ropes, and the like. The disadvantages of eddy current testing are also apparent. Firstly, the disturbance factors influencing the eddy current to generate variation are too much in the detection process, which is not beneficial to control. Moreover, the workpiece to be detected is required to be made of metal, and the image of the inner wall of the workpiece cannot be obtained in the detection process.
Laser speckle refers to various techniques that use superposition of speckle intensity patterns, one from the initial state of a rough object and one from after some form of displacement of the same object. The two patterns are recorded on the same coherent plate or the same detection array, thus forming a 'speckle pattern', and then a fringe pattern can be obtained by Fourier transform, namely, the coherent light illuminates a part of the speckle pattern and inspects the intensity distribution on the focal plane behind the lens. The method can effectively judge the flaws and cracks, can quickly obtain specific positions, and has high precision reaching submicron level. However, the speckle method obtains the surface characteristics by calculating statistical information, cannot restore the actual inner wall image, and is not favorable for visual observation.
The ultrasonic method is one of the main methods for nondestructive testing, and utilizes the mechanical wave property of ultrasonic waves to reflect when encountering a heterogeneous interface (defect and the like) with different acoustic impedances and convert the heterogeneous interface into electric pulses, thereby realizing the detection. The method has the disadvantages that coupling agents such as grease or water are needed in the measuring process, the method is easily interfered by the outside and influences can be generated on the pipe wall. The conventional ultrasonic method utilizes a probe to detect the workpiece point by point, has low efficiency, is influenced by human factors, and has subjective detection results. In addition, the method has a certain near-field blind area, easily causes missed detection on near-surface defects, and is not suitable for detecting thin-wall workpieces.
The manual visual inspection method uses a high-power magnifier to cooperate with an endoscopic part to scan the outer wall and the inner wall of the horn mouth of the catheter through human eyes, and then whether cracks, scratches and gouges exist is judged. The disadvantages of this approach are the following three:
(1) the efficiency is low: the whole process of the inspection needs the participation of inspectors, which wastes time and labor;
(2) the missed detection is more: the inner diameter of the part is small, the visual angle is small, the lighting is difficult, the outer wall inspection is difficult, and the problems of missing inspection or inconsistent individual judgment exist;
(3) tracing source is difficult: after the human eyes are checked, no data are stored, which is not beneficial to future checking and statistics.
The difficulty in solving the above problems and defects is: the inner and outer walls need to be detected simultaneously and the image data transmitted simultaneously.
The significance of solving the problems and the defects is as follows: through detecting pipe horn mouth inner and outer wall, obtain the complete image of part inner and outer wall, realize detecting pipe horn mouth surface quality, can be used for pipe horn mouth surface detection with digital measurement technique simultaneously, will detect data and digitize, and readability is good, can compress pipe horn mouth detection cycle, ensures assembly quality.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a method for detecting the surface defects of the inner wall and the outer wall of a horn mouth of an aircraft conduit.
The invention is realized in this way, a method for detecting the surface defect of the inner wall and the outer wall of the bell mouth of the aircraft guide pipe comprises the following steps:
step one, using an inner conical mirror, uniformly reflecting light guided by optical fibers to the outer wall of a workpiece by matching with a lamp strip, reflecting the light to a lens for imaging by the outer wall of the workpiece through the same light path, and combining the light striking of the outer wall and the light reflecting light path into a whole by the inner conical mirror;
step two, an endoscopic imaging head with a light source is adopted to be matched with a reflection system for illumination, so that large-range uniform irradiation is realized, and clear imaging of the curved surfaces of the inner wall and the outer wall is ensured;
thirdly, the endoscopic imaging head is imaged by the objective lens and transmitted to the CCD rake surface, then the CCD converts the optical signal into an electronic signal, the data is transmitted to the video endoscopic control group, and then the control group outputs the image to the computer;
and step four, training a target detection model by applying a deep learning method, and detecting defects in the image.
As the preferred embodiment of the invention, in the step one, the inner conical mirror completely reflects the image of the outer wall of the bell mouth, the image is clear without shielding and overexposure phenomena, and the angle and the size design meet the maximum utilization rate of the space.
As a preferred embodiment of the present invention, the handheld device of the inner conical mirror is composed of an endoscope portion, a self-processing inner conical mirror and a lamp strip, wherein the inner wall of the inner conical mirror is a highly reflective smooth workpiece, the endoscope portion is provided with an optical fiber light guide matching optical system, which can generate bright and uniform illumination light, the depth of endoscopic field and the angle of view are large, the simultaneous detection of the inner wall and the outer wall is realized, and the detection range and the detection precision are improved by selecting an endoscope head with a large field of view.
As a preferred embodiment of the invention, the angle of the inner conical mirror is designed according to the angle and the diameter of the horn mouth of the catheter.
In step two, the image is transferred by using the image transfer lens optical technology, and the light transmission illumination is provided by the optical fiber.
As a preferred embodiment of the invention, the size and the angle of the inner conical mirror are designed, and the outer wall of the bell mouth is designed relative to the optical path of the inner conical mirror.
The size and angle design, material selection and optical path design of the reflecting system are the preferred embodiments of the invention.
As a preferred embodiment of the invention, the angle and size of the inner cone is determined by the flare detail size d m and angle a, in terms of:
d＝13.55mm
a＝66°
the inner cone has dimensions of 23.64mm width, 23.64mm length and 6.75mm height.
The angle of the inner conical mirror is 30 degrees to the horizontal plane.
As a preferred embodiment of the present invention, in the fourth step, a deep learning algorithm fast-RCNN is adopted as a main algorithm for defect detection, a Resnet101 deep classification Network is used as a main structure for Feature extraction, and Feature Pyramid Network is used in combination to meet effective detection of defects of different sizes, and by means of a large amount of data labeling and optimization of algorithm parameters, the defects are accurately detected, detection in a complex environment is realized, and meanwhile, the area of the defects can be calculated by regression of a target detection box.
Another object of the present invention is to provide a system for implementing the method for detecting surface defects of an inner wall and an outer wall of an aircraft conduit bell mouth, the system for detecting surface defects of an inner wall and an outer wall of an aircraft conduit bell mouth comprises:
the self-made endoscopic imaging system comprises an objective lens, an ocular lens, a CCD imaging device and a signal transmission device;
the self-grinding reflection system is used for simultaneously imaging the inner wall and the outer wall and reflecting the image of the outer wall;
the inner conical mirror reflects the outer wall image;
and a lamp strip illuminating the outer wall portion.
The optical fiber light guide illumination is matched with a self-grinding reflection system to uniformly brighten the inner wall of the horn mouth of the catheter, the outer wall is uniformly brightened through the reflection of an inner cone mirror and the lamp strip, and the inner wall and the outer wall of the horn mouth of the catheter are imaged through an endoscopic probe; and (5) processing the image to detect the defects of the horn mouth of the catheter.
By combining all the technical schemes, the invention has the advantages and positive effects that:
(1) the detection requirements of the inner wall and the outer wall of the bell mouth of the conduit with the diameter of phi 4-phi 10 can be met, and the device is convenient to use, accurate and rapid in detection; has certain universality.
(2) The restored inner and outer wall images should reflect the surface morphology of the workpiece as truly as possible without bright spots, dark shadows and the like.
(3) The detection accuracy rate is 95%.
(4) The equipment can automatically mark cracks, scratches and collision defects, automatically count and complete confirmation and identification.
(5) The stability of the measuring system is good.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained from the drawings without creative efforts.
FIG. 1 is a detailed illustration of the inner and outer wall defect detection optical paths of the method for detecting the surface defects of the inner and outer walls of the horn mouth of the aircraft duct based on the optical path reflection of the conical mirror according to the embodiment of the present invention;
wherein, (a) is a mounting structure schematic diagram of an endoscope probe, and (b) is a structure schematic diagram of an endoscope reflector; (a) points a and b are indicated by outer walls; c. point d is indicated by the inner wall.
Fig. 2 is a schematic diagram of the design of the angle of the conical mirror in the method for detecting the surface defects of the inner wall and the outer wall of the aircraft conduit bell mouth based on the reflection of the optical path of the conical mirror provided by the embodiment of the invention.
Fig. 3 is a schematic view of an overall layout of a cone of the method for detecting defects on the inner and outer wall surfaces of a horn of an aircraft duct based on light path reflection of the cone mirror according to the embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a bell mouth of the method for detecting surface defects of the inner wall and the outer wall of the bell mouth of the aircraft conduit based on the light path reflection of the conical mirror provided by the embodiment of the invention.
FIG. 5 is a schematic diagram illustrating a principle of a home-made imaging head of a method for detecting surface defects of an inner wall and an outer wall of a horn mouth of an aircraft duct based on light path reflection of a conical mirror according to an embodiment of the present invention;
wherein, (a) is a schematic diagram of a self-made imaging head structure, and (b) is an image for detecting surface defects of the inner wall and the outer wall of a horn mouth of an aircraft conduit.
FIG. 6 shows inner wall scratches and crack defects of a method for detecting surface defects of inner and outer walls of a horn of an aircraft duct based on light path reflection of a cone mirror according to an embodiment of the present invention;
wherein, (a) is a defect diagram of scratching, gouging and cracking of the inner wall of the trumpet-shaped port of the phi 8 conduit, (b) is a defect diagram of scratching and cracking of the outer wall of the trumpet-shaped port of the phi 6 conduit, and (c) is a defect diagram of scratching and cracking of the inner wall and gouging of the outer wall of the trumpet-shaped port of the phi 6 conduit.
Fig. 7 is a flowchart of deep learning implementation of the method for detecting surface defects of the inner wall and the outer wall of the aircraft conduit bell mouth based on the light path reflection of the conical mirror provided by the embodiment of the invention.
Fig. 8 is a deep learning defect detection result diagram of the method for detecting the surface defects of the inner wall and the outer wall of the aircraft conduit bell mouth based on the light path reflection of the conical mirror provided by the embodiment of the invention.
Fig. 9 is a flowchart of a method for detecting surface defects of an inner wall and an outer wall of a horn of an aircraft duct based on light path reflection of a conical mirror according to an embodiment of the present invention.
In the figure: 1. an endoscopic probe; 2. an optical fiber; 3. a diffuse reflection plate; 4. self-processing a conical mirror; 5. a light strip; 6. an endoscopic mirror; 6-1, an upper diffuse reflection part; 6-2, a lower lens portion; 7. producing a product; 8. the rear end part of the eyepiece; 9. an image rotating lens group; 10. the tip of the CCD objective lens is imaged.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather should be construed as broadly as the present invention is capable of modification in various respects, all without departing from the spirit and scope of the present invention.
It will be understood that when an element is referred to as being "secured to" another element, it can be directly on the other element or intervening elements may also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. As used herein, the terms "vertical," "horizontal," "left," "right," and the like are for purposes of illustration only and are not intended to represent the only embodiments.
Unless defined otherwise, 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 invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
In view of the problems in the prior art, the invention provides a method for detecting defects on the inner wall and the outer wall of a horn mouth of an aircraft duct, and the invention is described in detail with reference to the accompanying drawings.
As shown in fig. 1 (a): this aircraft pipe bell mouth inner and outer wall surface defect detecting system includes: the optical fiber light guide 1(a) -2 is matched with a self-grinding reflection system for illumination; the inner conical mirrors 1(a) -4 are used for uniformly reflecting the light to the outer wall of the workpiece, and the outer wall of the workpiece can also reflect the light to the lens through the same light path; the self-made endoscopic imaging head 1(a) -1 is imaged by the objective lens and transmitted to the CCD rake surface, then the CCD converts the optical signal into an electronic signal, the data is transmitted to the video endoscopic control group, and the image is output to the computer by the control group. And after the images are collected, detecting scratches, gouges and cracks by using a deep learning algorithm.
The method for detecting the surface defects of the inner wall and the outer wall of the horn mouth of the aircraft guide pipe comprises the following steps:
(1) the inner conical mirrors 1(a) -4 are innovatively used, light can be uniformly reflected to the outer wall of the workpiece, and the outer wall of the workpiece can reflect the light to the lens 1(a) -1 through the same light path. The inner cone lens can combine the outer wall polishing and light reflection light paths into a whole, and the purpose of detecting the outer wall of the workpiece is achieved simply and effectively.
(2) The self-made endoscopic imaging head (with a light source) 1(a) -1 is matched with a self-grinding reflection system to illuminate, so that light is prevented from being concentrated in the center of the inner wall to be overexposed, large-range uniform irradiation is realized, and clear imaging of the curved surfaces of the inner wall and the outer wall is ensured. Endoscopic imaging leads light by an optical fiber, and has high light flux, quick lightening and uniform illuminance matched with a self-grinding reflection system; the novel multifunctional pen has the advantages of small volume, large field angle, convenience in holding, energy conservation and environmental friendliness.
(3) The self-made endoscopic imaging head 1(a) -1 is imaged by the objective lens and transmitted to the CCD rake surface, then the CCD converts the optical signal into an electronic signal, the data is transmitted to the video endoscopic control group, and the image is output to the computer by the control group. The image is transferred by using the optical technology of a relay lens, and light transmission illumination is provided by an optical fiber.
(4) And (3) training a target detection model by applying a deep learning method, and detecting defects in the image.
The inner cone mirror must completely reflect images of the outer wall of the horn mouth, imaging is clear, shielding and overexposure phenomena do not exist, and the angle and size design meets the maximum space utilization rate. The design concept of the size and angle of the self-grinding reflection system is as follows in FIG. 1 (a):
the handheld device consists of an endoscope part, a self-processing conical mirror and a lamp strip. Because the inner wall is a high-reflection smooth workpiece, the endoscopic part can generate bright and uniform illumination light by matching light guide with the optical fiber and the self-grinding optical system, the depth of field and the angle of field of view of the endoscope are large, the simultaneous detection of the inner wall and the outer wall can be realized, and the detection range and the detection precision can be improved by selecting the endoscopic lens with a large field of view. Points a and b are bell mouth outer wall points; points c and d are inner wall points of the bell mouth; the relatively low a point on the outer wall is reflected by the inner conical surface with a smaller diameter into the lens, so that the ring on the imaging plane is like a relatively small ring a'. And the point b with a relatively higher position on the outer wall is reflected into the lens by the conical surface with a larger diameter, so that the circular ring on the imaging surface is like a relatively larger circular ring b'. The light of point c on the inner wall enters the lens directly, so the circle on the imaging surface is like a smaller circle c'. The light of point d on the inner wall directly enters the lens, so the circular ring on the imaging surface is like a larger circular ring d'.
The technical idea of designing the size and the angle of the inner conical mirror is shown in fig. 2:
the measurement scheme innovatively employs an inner conical mirror. Through interior conical mirror, can reflect the light cooperation lamp area of optic fibre leaded light to the work piece outer wall uniformly. The outer wall of the workpiece can also reflect light to the lens for imaging through the same light path. The inner cone reflector can combine the outer wall polishing and light reflecting light paths into a whole, and the purpose of simultaneously detecting the inner wall and the outer wall of the workpiece is simply and effectively achieved. The endoscope reflector consists of two parts, the upper part is a diffuse reflection part, and the lower part is a lens part. The endoscope reflecting mirror is fixedly installed with the endoscope probe through the adapter ring. As shown in fig. 2: the angle and size of the inner cone is determined by the flare detail size d m and angle a, according to:
d＝13.5mm
a＝66°
the size of the inner cone mirror is 23.64mm in width, 23.64mm in length and 6.75mm in height. (the final size is determined by calculation and experiment according to the type and size of the part model).
This scheme is according to actual demand, will realize the accurate detection to pipe horn mouth inner and outer wall defect, and equipment needs to reach following technical parameter.
(1) The detection requirements of the inner wall and the outer wall of the bell mouth of the conduit with the diameter of phi 4-phi 10 can be met, and the device is convenient to use, accurate and rapid in detection; has certain universality.
(2) The restored inner and outer wall images should reflect the surface morphology of the workpiece as truly as possible without bright spots, dark shadows and the like.
(3) The detection accuracy is more than or equal to 95 percent.
(4) The equipment can automatically mark cracks, scratches and collision defects, automatically count and complete confirmation and identification.
(5) The stability of the measuring system is good.
In order to realize the detection requirements, an optical inner and outer wall visual detection method based on a conical mirror is provided by combining the theoretical and practical basis of a laboratory in the defect detection direction and utilizing the principle of optical reflection to obtain images of the inner and outer walls. The system can effectively acquire the images of the inner wall and the outer wall of the workpiece, is simple to operate, is beneficial to image storage, has high resolution and repeatability, transmits the images into the algorithm server by the equipment in a wired mode and a wireless mode, and simultaneously displays the results of defective products on the screen of the equipment. The equipment composition is shown in 5:
(1) the inner conical mirror is innovatively used, light can be uniformly reflected to the outer wall of the workpiece, and the outer wall of the workpiece can reflect the light to the lens through the same light path. The inner cone lens can combine the outer wall polishing and light reflection light paths into a whole, and the purpose of detecting the outer wall of the workpiece is achieved simply and effectively.
(2) The self-made endoscopic imaging head (with a light source) is matched with the self-grinding reflection system for illumination, so that light is prevented from being concentrated in the center of the inner wall for overexposure, large-range uniform irradiation is realized, and clear imaging of the curved surfaces of the inner wall and the outer wall is ensured. Endoscopic imaging leads light by an optical fiber, and has high light flux, quick lightening and uniform illuminance matched with a self-grinding reflection system; the novel multifunctional pen has the advantages of small volume, large field angle, convenience in holding, energy conservation and environmental friendliness.
(3) The self-made endoscopic imaging head is imaged by the objective lens and transmitted to the CCD rake surface, then the CCD converts the optical signal into an electronic signal, the data is transmitted to the video endoscopic control group, and the image is output to the computer by the control group. The image is transferred by using the optical technology of a relay lens, and light transmission illumination is provided by an optical fiber.
As shown in fig. 1: the handheld device consists of an endoscopic part, a self-processing conical mirror and a lamp strip. Because the inner wall is a high-reflection smooth workpiece, the endoscopic part can generate bright and uniform illumination light by matching light guide with the optical fiber and the self-grinding optical system, the depth of field and the angle of field of view of the endoscope are large, the simultaneous detection of the inner wall and the outer wall can be realized, and the detection range and the detection precision can be improved by selecting the endoscopic lens with a large field of view.
In fig. 6, (a) is a drawing of defects of scratching, gouging and cracking of the inner wall of the trumpet-shaped port of the phi 8 catheter, (b) is a drawing of defects of scratching and cracking of the inner wall of the trumpet-shaped port of the phi 6 catheter, and (c) is a drawing of defects of scratching and cracking of the inner wall of the trumpet-shaped port of the phi 6 catheter. The defect is clear and has no shielding, and the light is uniform.
A deep learning algorithm fast-RCNN is adopted as a main algorithm of defect detection, a Resnet101 deep classification network is used as a main structure of feature extraction, and FPN (feature Pyramid network) is matched for use, so that the defects with different sizes can be effectively detected. The defects are accurately detected through a large amount of data labeling and adjustment and optimization of algorithm parameters, detection under a complex environment is achieved, and meanwhile the areas of the defects can be calculated through regression of a target detection frame. Regarding the implementation flow of deep learning, as shown in fig. 7:
and analyzing according to a part of photos collected by the sample piece in the earlier stage, and effectively detecting cracks and scratches as shown in fig. 8, wherein the framed area is a defect existence area. In the early stage, the horn mouth images need to be collected in a large scale for deep learning, marking and training.
The defect detection equipment for the inner wall and the outer wall of the horn mouth of the catheter consists of a self-made endoscopic imaging system, a self-grinding reflection system, an inner conical mirror and a lamp strip, and has the functions of: the optical fiber light guide illumination is matched with a self-grinding reflection system to uniformly brighten the inner wall of the horn mouth of the catheter, the outer wall is uniformly brightened through the reflection of an inner cone mirror and the lamp strip, and the inner wall and the outer wall of the horn mouth of the catheter are imaged through an endoscopic probe; and (5) processing the image to detect the defects of the horn mouth of the catheter.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure should be limited only by the attached claims.
Claims (7)
1. The method for detecting the surface defects of the inner wall and the outer wall of the horn mouth of the aircraft guide pipe is characterized by comprising the following steps of:
step one, using an inner conical mirror, uniformly reflecting light guided by optical fibers to the outer wall of a workpiece by matching with a lamp strip, reflecting the light to a lens for imaging by the outer wall of the workpiece through the same light path, and combining the light striking of the outer wall and the light reflecting light path into a whole by the inner conical mirror; the handheld device of the inner conical mirror consists of an endoscope part, a self-processing conical mirror and a lamp strip; the inner wall of the inner conical mirror is a high-light-reflection smooth workpiece, the endoscopic part can generate bright and uniform illumination light by virtue of the optical fiber light guide matching optical system, the endoscopic depth of field and the field angle are large, the simultaneous detection of the inner wall and the outer wall is realized, and the detection range and the detection precision are improved by selecting an endoscopic lens with a large field of view; the angle of the inner conical mirror is designed according to the angle and the diameter of the horn mouth of the conduit;
step two, an endoscopic imaging head with a light source is adopted to be matched with a reflection system for illumination, so that large-range uniform irradiation is realized, and clear imaging of the curved surfaces of the inner wall and the outer wall is ensured;
thirdly, the endoscopic imaging head is imaged by the objective lens and transmitted to the CCD rake surface, then the CCD converts the optical signal into an electronic signal, the data is transmitted to the video endoscopic control group, and then the control group outputs the image to the computer;
and step four, training a target detection model by applying a deep learning method, and detecting defects in the image.
2. The method for detecting the surface defects of the inner wall and the outer wall of the aircraft conduit bell mouth as claimed in claim 1, wherein in the step one, the inner conical mirror completely reflects the image of the outer wall of the bell mouth, the image is clear without shielding and overexposure, and the angle and the size are designed to meet the maximum utilization rate of space.
3. The method as claimed in claim 1, wherein in step two, the image is transmitted by using a relay lens optical technique, and the light transmission illumination is provided by the optical fiber.
4. The method of claim 1, wherein the size and angle of the inner conical mirror are designed such that the outer wall of the flare is designed relative to the optical path of the inner conical mirror.
5. The method for detecting the surface defects of the inner wall and the outer wall of the aircraft conduit bell mouth as claimed in claim 1, wherein the angle and the size of the inner conical mirror are determined by the size d and the angle a of the bell mouth part according to the following steps:
d＝13.55mm
a＝66°
the size of the inner cone mirror is 23.64mm in width, 23.64mm in length and 6.75mm in height.
6. The method for detecting the surface defects of the inner wall and the outer wall of the horn mouth of the aircraft duct according to claim 1, wherein in the fourth step, a deep learning algorithm fast-RCNN is adopted as a defect detection algorithm, a Resnet101 deep classification Network is adopted as a main structure for Feature extraction, and a Feature Pyramid Network is used in a matching manner, so that the defects with different sizes can be effectively detected, the defects can be accurately detected through a large amount of data labeling and optimization of algorithm parameters, the detection under the complex environment is realized, and meanwhile, the area of the defects is calculated through regression of a target detection box.
7. A system for implementing the method for detecting the surface defects of the inner wall and the outer wall of the aircraft conduit bell mouth according to any one of claims 1 to 6, wherein the system for detecting the surface defects of the inner wall and the outer wall of the aircraft conduit bell mouth comprises:
the self-made endoscopic imaging system consists of an objective lens, an ocular lens and a CCD imaging and signal transmission device;
the self-grinding reflection system is used for simultaneously imaging the inner wall and the outer wall and reflecting the image of the outer wall;
the inner conical mirror reflects the outer wall image;
a light strip illuminating the outer wall portion;
the optical fiber light guide illumination is matched with a self-grinding reflection system to uniformly brighten the inner wall of the horn mouth of the catheter, the outer wall is uniformly brightened through the reflection of an inner cone mirror and the lamp strip, and the inner wall and the outer wall of the horn mouth of the catheter are imaged through an endoscopic probe; and (5) processing the image to detect the defects of the horn mouth of the catheter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011225991.5A CN112345551B (en) | 2020-11-05 | 2020-11-05 | Method and system for detecting surface defects of inner wall and outer wall of aircraft conduit bell mouth |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011225991.5A CN112345551B (en) | 2020-11-05 | 2020-11-05 | Method and system for detecting surface defects of inner wall and outer wall of aircraft conduit bell mouth |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112345551A CN112345551A (en) | 2021-02-09 |
| CN112345551B true CN112345551B (en) | 2021-07-30 |
Family
ID=74429902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011225991.5A Active CN112345551B (en) | 2020-11-05 | 2020-11-05 | Method and system for detecting surface defects of inner wall and outer wall of aircraft conduit bell mouth |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112345551B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113267452A (en) * | 2021-06-11 | 2021-08-17 | 菲特（天津）检测技术有限公司 | Engine cylinder surface defect detection method and system based on machine vision |
| CN114018948A (en) * | 2021-11-11 | 2022-02-08 | 合肥君信电子科技有限公司 | Detection probe for endoscope |
| CN113984769B (en) * | 2021-12-27 | 2022-03-22 | 江苏美克医学技术有限公司 | Light ray homogenization treatment based digital reading device for detection result of kit/card |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105277563A (en) * | 2015-11-19 | 2016-01-27 | 中国民航大学 | Prevailing visibility automatic observation method based on hemisphere mirror principle |
| CN107049214A (en) * | 2017-04-27 | 2017-08-18 | 苏州双威医疗器械科技有限公司 | Automatic focusing microscopy endoscopic fluoroscopic imaging systems |
| KR20180035484A (en) * | 2016-09-29 | 2018-04-06 | 주식회사 제이이노텍 | Apparatus and method for inspection of welding state of cylindrical battery |
| EP2813804B1 (en) * | 2012-02-09 | 2019-11-13 | IHI Corporation | Inside-diameter measurement device |
Family Cites Families (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| BE875482A (en) * | 1979-04-11 | 1979-10-11 | Chandesais Jean L J | IMPROVED RING LIGHT FOR MICROSBUIES |
| CN201269491Y (en) * | 2008-10-14 | 2009-07-08 | 上海创波光电科技有限公司 | Illumination device used for detecting cylindrical surface defect |
| CN101504376A (en) * | 2009-03-11 | 2009-08-12 | 张勇 | Optical detection method and apparatus for spherical object surface |
| CN202649134U (en) * | 2012-06-08 | 2013-01-02 | 北京大恒图像视觉有限公司 | Bottle cap quality detection device |
| CN203465208U (en) * | 2013-09-27 | 2014-03-05 | 天津欣维检测技术有限公司 | Nondestructive detection device for appearance of inner wall of straight pipe |
| CN103743756A (en) * | 2014-01-20 | 2014-04-23 | 清华大学 | Online detecting device and method for inner wall of vessel |
| CN105105699B (en) * | 2015-09-06 | 2017-03-22 | 深圳英美达医疗技术有限公司 | Fiber optic endoscope |
| CN205262996U (en) * | 2015-11-27 | 2016-05-25 | 云南电网有限责任公司昭通供电局 | Reactor internal surface detection device |
| CN206114481U (en) * | 2016-08-28 | 2017-04-19 | 郑州赛斯工业技术转化研究院有限公司 | Test device that water permeability that absorbs water recycled concrete detected |
| JP2019117139A (en) * | 2017-12-27 | 2019-07-18 | 株式会社ケージェーケー | Bearing inspection method and inspection device |
| FR3076619B1 (en) * | 2018-01-05 | 2020-01-24 | Tiama | METHOD, DEVICE AND INSPECTION LINE FOR DETERMINING THREE-DIMENSIONAL GEOMETRY OF A CONTAINER RING SURFACE |
| CN208187367U (en) * | 2018-05-10 | 2018-12-04 | 徐州市宏达土木工程试验室有限责任公司 | A kind of novel lossless detection device |
| CN109030492A (en) * | 2018-05-25 | 2018-12-18 | 芜湖天航装备技术有限公司 | A kind of conduit horn mouth surface quality detection device |
| CN208751997U (en) * | 2018-08-10 | 2019-04-16 | 苏州红隼机电科技有限公司 | Defect detecting device inside and outside a kind of seamless steel pipe |
| CN109459447B (en) * | 2018-11-23 | 2023-09-12 | 菲特（天津）检测技术有限公司 | Luminous light source structure, cylindrical workpiece detection device and detection method |
| CN109613020B (en) * | 2019-01-28 | 2023-06-09 | 重庆中烟工业有限责任公司 | Clamping type cigarette hole size detection device |
| CN210894142U (en) * | 2019-07-02 | 2020-06-30 | 东莞创视自动化科技有限公司 | Multi-view optical imaging device |
| CN210773850U (en) * | 2019-10-30 | 2020-06-16 | 上海航翼高新技术发展研究院有限公司 | Aircraft metal pipe horn mouth angle detects instrument |
| CN111650203B (en) * | 2020-04-24 | 2023-03-28 | 南京理工大学 | Method for measuring defects on inner surface of microsphere |
-
2020
- 2020-11-05 CN CN202011225991.5A patent/CN112345551B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2813804B1 (en) * | 2012-02-09 | 2019-11-13 | IHI Corporation | Inside-diameter measurement device |
| CN105277563A (en) * | 2015-11-19 | 2016-01-27 | 中国民航大学 | Prevailing visibility automatic observation method based on hemisphere mirror principle |
| KR20180035484A (en) * | 2016-09-29 | 2018-04-06 | 주식회사 제이이노텍 | Apparatus and method for inspection of welding state of cylindrical battery |
| CN107049214A (en) * | 2017-04-27 | 2017-08-18 | 苏州双威医疗器械科技有限公司 | Automatic focusing microscopy endoscopic fluoroscopic imaging systems |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112345551A (en) | 2021-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112345551B (en) | Method and system for detecting surface defects of inner wall and outer wall of aircraft conduit bell mouth | |
| US10782248B2 (en) | Automatic detection device and method for detecting surface detects of large caliber cambered optical element | |
| Zhao | Review of non-destructive testing methods for defect detection of ceramics | |
| CN107255454B (en) | A kind of multiple dimensioned multi-functional strain measurement system of superhigh temperature based on ultraviolet imagery DIC and measurement method | |
| CN100342207C (en) | Detector for three-dimensional appearance of micro-member through-hole inner surface and its marking and using method | |
| CN202453298U (en) | Surface micro-defect detecting device based on laser scattering | |
| CN108332689A (en) | A kind of optical measuring system and method for detection surface roughness and surface damage | |
| CN101382422A (en) | Internal contour outline automatic detection system for pipe-shaped parts | |
| CN105157584B (en) | A kind of on-line measurement device and method of non-contact object thickness | |
| US10207297B2 (en) | Method and system for inspecting a manufactured part at an inspection station | |
| CN111323480B (en) | Handheld automatic focusing laser ultrasonic nondestructive testing system | |
| CN108759664B (en) | Nuclear fuel rod size detection method and system based on telecentric optics | |
| CN102507596B (en) | Optical element surface defect detecting system based on active laser beam scanning | |
| CN111288902B (en) | Double-field-of-view optical coherence tomography imaging system and material thickness detection method | |
| WO2010084314A1 (en) | Optical measuring method and system | |
| CN109799192A (en) | A kind of non-contact laser ultrasonic nondestructive testing and method for arbitrary surface | |
| CN108593710B (en) | Thermal imaging detection system and method for surface defects of high-reflectivity material | |
| Zhou et al. | Defect detection for highly reflective rotary surfaces: an overview | |
| CN112881403A (en) | Hot rolling strip steel surface defect detection device | |
| Yoshizawa et al. | Development of an inner profile measurement instrument using a ring beam device | |
| CN201034557Y (en) | Cylinder inside and outside wall machine precision on-line image-forming testing apparatus | |
| CN113686903B (en) | Optical element defect detection system and detection method | |
| CN2562183Y (en) | Fine tube internal surface detector | |
| CN115389620A (en) | Laser ultrasonic nondestructive testing device and method for high-temperature load in-situ detection | |
| CN113933026B (en) | Lens surface flaw detection device and method based on transmission and reflection mixed illumination |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210209 Assignee: Lizhi energy saving Co.,Ltd. Assignor: FITOW (TIANJIN) DETECTION TECHNOLOGY CO.,LTD. Contract record no.: X2021980011398 Denomination of invention: A surface defect detection method and system for inner and outer wall of aircraft conduit bell mouth Granted publication date: 20210730 License type: Exclusive License Record date: 20211028 |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: Lizhi energy saving Co.,Ltd. Assignor: FITOW (TIANJIN) DETECTION TECHNOLOGY CO.,LTD. Contract record no.: X2021980011398 Date of cancellation: 20230712 |