CN105866241A - Machine-vision self-analyzing flaw detection device for shaft parts and machine-vision self-analyzing flaw detection method - Google Patents

Abstract

The invention relates to a flaw detection device and method for machine vision self-analysis shaft parts. The flaw detection device includes a housing, a guide rail, a screw, a nut, a first stepping motor, a second stepping motor, a guide rail, a slider, and a first support block, a second support block and a wheel pair, the guide rail is located in the middle of the housing, a slide block is installed on the guide rail, the slide block is connected with the second support block, and the wheel pair is installed on the second support block, Shaft parts are placed on the wheel set; the screw is located above the guide rail, the nut is threadedly connected to the screw, the first support block is fixedly connected to the nut, and a CCD camera, Ultraviolet lamp and magnetic suspension nozzle: the two ends of the shaft parts are respectively provided with magnetizers, and the ends of the magnetizers are provided with yokes. The patent of the invention can use the CCD camera to collect scar images, and perform self-processing to reduce ultraviolet damage, and accelerate scar recognition processing to speed up recognition efficiency.

Description

Machine vision self-analysis shaft parts flaw detection device and method

technical field

The invention relates to flaw detection equipment, more specifically, to a flaw detection device for machine vision self-analysis shaft parts.

Background technique

Fluorescent magnetic particle inspection is a commonly used non-destructive testing method, and it is the preferred method for detecting surface and near-surface defects of steel parts. In China, technicians rely on human eyes combined with personal experience to identify defects displayed by phosphors, which is less efficient and less accurate.

With the increasing maturity of digital image processing technology, it is the future trend of magnetic particle flaw detection technology to use CCD camera instead of human eyes to complete the complicated work of identifying cracks. CCD technology has been put into use in China, but the real detection and analysis still has to be done manually.

Contents of the invention

The technical problem to be solved by the present invention is to provide a machine vision self-analysis shaft parts flaw detection device and method, which can reduce the amount of manual labor and improve efficiency and accuracy.

The technical solution adopted by the present invention to solve the technical problem is: to construct a machine vision self-analysis shaft parts flaw detection device, including a housing, the housing is equipped with a guide rail, a screw, a nut, a first stepping motor, a second Two stepper motors, guide rails, sliders, first support blocks, second support blocks and wheel sets;

The guide rail is located in the middle of the housing, a slider is installed on the guide rail, the slider is connected to the second support block, the wheel set is installed on the second support block, and the shaft parts are placed on the wheel set;

The screw mandrel is located above the guide rail, the nut is threadedly connected with the screw mandrel, the first support block is fixedly connected with the nut, and a CCD camera, an ultraviolet lamp and a magnetic suspension nozzle are arranged below the first support block:

The first stepping motor drives the screw to rotate, and the second stepping motor drives the shaft parts to rotate. The two ends of the shaft parts are respectively provided with magnetizers, and the ends of the magnetizers are provided with magnetizers. yoke.

In the above solution, a bottom box is provided at the bottom of the housing, and the bottom box is connected to the magnetic suspension nozzle through a water pump and a conduit.

In the above solution, the second stepper motor is fixed at the bottom of the housing, the end of the shaft part is connected to the first pulley, the output shaft of the second stepper motor is connected to the second pulley, and the first pulley and the second pulley pass through the belt connect.

The present invention also provides a flaw detection method of the above-mentioned machine vision self-analysis shaft flaw detection device, which includes the following steps:

S1. Install the shaft parts on the wheel set, the shaft parts are magnetized by the yoke and the magnetizer, the first stepping motor and the second stepping motor rotate, the shaft parts rotate, the magnetic suspension nozzle moves horizontally, and the Fluorescent magnetic particle suspension is sprayed onto the surface of shaft parts;

S2. Shaft parts will generate leakage magnetic field at defective cracks. After spraying fluorescent magnetic particle suspension, the leakage magnetic field will absorb fluorescent magnetic powder, and the fluorescent magnetic powder will turn green under the irradiation of ultraviolet light, and the CCD camera will collect the fluorescence;

S3. By performing binarization, median filtering, and defect identification processing on the collected images, some interference information in the images is eliminated, and only the final shape, size and quantity of cracks are displayed on the computer.

Implementing the machine vision self-analysis shaft parts flaw detection device and method of the present invention has the following beneficial effects:

The invention uses a CCD to collect fluorescent images, and imports the collected images into a computer image recognition system for analysis and processing. Identify the scars and remove some interference information, identify the size of the scars and record the number of scars, analyze whether the shaft parts can continue to work, and recommend repair methods, and push the above information to the computer terminal of the flaw detector. The flaw detector only needs to carry out Confirmation work improves the accuracy and efficiency of flaw detection.

Description of drawings

The present invention will be further described below in conjunction with accompanying drawing and embodiment, in the accompanying drawing:

Fig. 1 is a front view of the patented machine vision self-analysis shaft parts flaw detection device of the present invention;

Fig. 2 is a flow chart of the patented machine vision self-analysis method for flaw detection of shaft parts of the present invention;

Fig. 3 is an image processing flow chart of the patented machine vision self-analysis method for flaw detection of shaft parts of the present invention.

detailed description

In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific implementation manners of the present invention will now be described in detail with reference to the accompanying drawings.

Please refer to accompanying drawings 1 and 2, the machine vision self-analysis shaft parts flaw detection device of the present invention includes a CCD camera 1, a shaft part 2, a magnetizer 3, a v-belt 4, a screw mandrel 5, a nut 6, First pulley 7, second pulley 8, first stepping motor 9, second stepping motor 10, guide rail 11, slider 12, wheel pair 13, yoke 14, ultraviolet lamp 15, magnetic suspension nozzle 16, conduit 17. Water pump 18, bottom box 19, casing 20, first support block 21, second support block 22.

The screw rod 5 is arranged in the casing 20 and is located at the top center of the casing 20 . The CCD camera 1, the ultraviolet lamp 15, and the magnetic suspension nozzle 16 are fixed on the first supporting block 21 through bolts and nuts. The ultraviolet 15 lamp is arranged in the middle of the first support block 21, the CCD camera 1 is arranged on the left side of the ultraviolet lamp 15, and the magnetic suspension nozzle 16 is fixed on the right side of the CCD camera 1. The first supporting block 21 on which the CCD camera 1 , the ultraviolet lamp 15 , and the magnetic suspension nozzle 16 are installed is affixed together with the nut 6 , and the nut 6 and the screw mandrel 5 are connected by threads. The screw mandrel 5 is connected with the first stepping motor 9 through a taper sleeve, and the first stepping motor 9 is located at the upper left in the housing.

The first pulley 7 is arranged on one side of the inner wall of the casing 20 and is located on the right side of the magnetizer 3 . The shaft part 2 is connected with the first pulley 7 through a taper sleeve, the shaft part 2 and the first pulley 7 rotate coaxially, and the first pulley 7 is connected with the second pulley 8 through the V-belt 4 . The second pulley 8 is connected with the second stepper motor 10 through a tapered sleeve, the second pulley 8 is located directly below the first pulley 7 , and the second stepper motor 10 is located on the left side of the second pulley 8 . The guide rail 11 is horizontally arranged in the middle part of the housing 20, and a pair of wheels 13 are respectively fixed on the shafts of a pair of second support blocks 22, and the shaft parts 2 are placed flat on the pair of wheels 13, and a pair of second wheels The supporting blocks 22 are respectively fixed on the pair of sliding blocks 12 by screws. A pair of sliding blocks 12 on which the second support block 22 is installed are respectively in sliding contact with the guide rails 11 .

A pair of yokes 14 and a magnetizer 3 are fixed on the left and right inner walls of the casing 20 by bolts and nuts, and the shaft part 2 is located between the pair of yokes 14 . The bottom box 19 is located at the middle and lower part of the housing, the water pump 18 is located at the lower right side of the housing 20 , and the conduit 17 connects the bottom box 19 with the water pump 18 and the magnetic suspension nozzle 16 .

After the shaft part 2 is magnetized by the yoke 14 and the magnetizer 3 and sprayed with the fluorescent magnetic powder suspension by the magnetic suspension nozzle 16, a leakage magnetic field will be generated at the defective crack. After spraying the fluorescent magnetic powder suspension, the leakage magnetic field will Adsorb fluorescent magnetic powder, and fluorescent magnetic powder is green under the irradiation of ultraviolet light. Use CCD imaging technology to collect fluorescence, and then perform binarization, median filtering, and defect identification processing on the collected images, which can eliminate some interference information in the image, and only display the final crack shape, size and quantity on the computer , and analyze whether the shaft parts can continue to work, and recommend repair methods.

A stepper motor and its driver 10 are arranged on the upper left in the housing 20. The stepper motor and its driver 10 are connected with one end of the screw mandrel 5 through a taper sleeve. The stepper motor and its driver 10 drive the screw mandrel 5 to rotate, and the nut 6 is in the The screw rod 5 moves linearly, the first support block 21 and the nut 6 are solidified together, the first support block 21 drives the CCD camera 1, the ultraviolet lamp 15, and the magnetic suspension nozzle 16 to move back and forth, the ultraviolet lamp 15 is turned on to provide a light source, and the magnetic The suspension nozzle 16 uses the rotation of the shaft part 2 and its own back and forth motion to evenly spray each part of the shaft part 2, and the CCD camera 1 uses the rotation of the shaft part 2 and its own back and forth motion to spray each part of the shaft part 2. The site is scanned and images are collected.

CCD camera 1, ultraviolet lamp 15, magnetic suspension nozzle 16 are installed on the first support block 21 by bolt nut, first support block 21 and nut 6 are consolidated together, and nut moves linearly along screw mandrel, CCD camera 1, ultraviolet ray The lamp 15 and the magnetic suspension nozzle 16 move synchronously.

The bottom box 19 collects the dropped magnetic suspension, and the water pump 18 pumps the magnetic suspension in the bottom box 19 back to the magnetic suspension nozzle 16 through the conduit 17 for spraying.

The present invention also provides a flaw detection method of the above machine vision self-analysis shaft parts flaw detection device, comprising the following steps:

S1. Install the shaft parts on the wheel set, the shaft parts are magnetized by the yoke and the magnetizer, the first stepping motor and the second stepping motor rotate, the shaft parts rotate, the magnetic suspension nozzle moves horizontally, and the Fluorescent magnetic particle suspension is sprayed onto the surface of shaft parts.

S2. Shaft parts will generate leakage magnetic field at defective cracks. After spraying fluorescent magnetic particle suspension, the leakage magnetic field will absorb fluorescent magnetic powder, and the fluorescent magnetic powder will turn green under the irradiation of ultraviolet light, and the CCD camera will collect the fluorescence.

S3, the image collected by the CCD camera is sent to the computer for image processing, first the grayscale image collected is binarized, and each point of the grayscale image is represented by 0 to 255. The brightness of this point, 0 is pure black, 255 is pure white. Binarization is to set a threshold, change the points whose gray value is less than the threshold to 0, and the points greater than the threshold to 255, then the image becomes composed of points with gray values of 0 and 255. Since this device uses fluorescent magnetic particle detection, the obtained image is bright and dark, and the threshold is set to 255, which can remove useless information and directly separate the target. There are still many noise points in the binarized image, and further processing is required to eliminate the interference factors. For the design of this device, the median filter is to replace the value of a point in the digital image with the median value of each point value in a neighborhood of the point, so that the surrounding pixel values are close to the real value, thereby eliminating isolated noise points. The collected images mainly appear "salt and pepper noise", manifested as scattered white points, which should be processed by median filtering. Considering that it does not affect the target, this system uses the median filter of the 3*3 area, that is, selects a total of 9 pixel points in the 3*3 area on the image, sorts them according to the gray value, and then replaces each point with the median value value to eliminate noise. Then use defect identification, that is, use an algorithm to calculate the ratio of the area of the defect area (the white area in the figure) to the area of the acquisition area (the crankshaft area in the figure), so that the scars of the measured shaft parts can be more intuitively displayed. At the same time, the number of white areas on the crankshaft is recorded by the counting module, and the number of scars can be obtained. By analyzing and comparing the shape, size and number of scars in the collected images with similar situations in the database, it is analyzed whether the shaft part can continue to be used, and a repair method is recommended, and the above information is pushed to the computer of the flaw detector on the terminal.

Based on the traditional method, the present invention uses CCD to collect fluorescent images, and imports the collected images into a computer image recognition system for analysis and processing, to identify scars and remove part of the interference information, to identify the size of scars and record the number of scars, and to analyze the shaft parts Whether it can continue to work, and recommended repair methods, and the above information is pushed to the computer terminal of the flaw detector, and the flaw detector only needs to confirm the work, which improves the accuracy and efficiency of flaw detection.

Embodiments of the present invention have been described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementations, and the above-mentioned specific implementations are only illustrative, rather than restrictive, and those of ordinary skill in the art will Under the enlightenment of the present invention, many forms can also be made without departing from the gist of the present invention and the protection scope of the claims, and these all belong to the protection of the present invention.

Claims (4)

1. A machine vision self-analysis shaft parts flaw detection device is characterized in that it comprises a housing, and the housing is provided with a guide rail, a screw mandrel, a nut, a first stepping motor, a second stepping motor, a guide rail, a slide block, the first support block, the second support block and the wheelset; The guide rail is located in the middle of the housing, a slider is installed on the guide rail, the slider is connected to the second support block, the wheel set is installed on the second support block, and the shaft parts are placed on the wheel set; The screw mandrel is located above the guide rail, the nut is threadedly connected with the screw mandrel, the first support block is fixedly connected with the nut, and a CCD camera, an ultraviolet lamp and a magnetic suspension nozzle are arranged below the first support block: The first stepping motor drives the screw to rotate, and the second stepping motor drives the shaft parts to rotate. The two ends of the shaft parts are respectively provided with magnetizers, and the ends of the magnetizers are provided with magnetizers. yoke. 2. The machine vision self-analyzing flaw detection device for shaft parts according to claim 1, wherein a bottom box is provided at the bottom of the housing, and the bottom box is connected to the magnetic suspension nozzle through a water pump and a conduit. 3. The flaw detection device for machine vision self-analysis shaft parts according to claim 1, wherein the second stepper motor is fixed on the bottom of the housing, and the end of the shaft parts is connected to the first pulley, and the second step The output shaft of the feed motor is connected with the second pulley, and the first pulley and the second pulley are connected by a belt. 4. A flaw detection method utilizing the machine vision self-analysis shaft part flaw detection device according to claim 1, characterized in that it comprises the following steps: S1. Install the shaft parts on the wheel set, the shaft parts are magnetized by the yoke and the magnetizer, the first stepping motor and the second stepping motor rotate, the shaft parts rotate, the magnetic suspension nozzle moves horizontally, and the Fluorescent magnetic particle suspension is sprayed onto the surface of shaft parts; S2. Shaft parts will generate leakage magnetic field at defective cracks. After spraying fluorescent magnetic particle suspension, the leakage magnetic field will absorb fluorescent magnetic powder, and the fluorescent magnetic powder will turn green under the irradiation of ultraviolet light, and the CCD camera will collect the fluorescence; S3. By performing binarization, median filtering, and defect identification processing on the collected images, some interference information in the images is eliminated, and only the final shape, size and quantity of cracks are displayed on the computer.