Intelligent identification method for magnetic powder inspection
A technology of intelligent identification and magnetic particle inspection, applied in the fields of material magnetic variables, image analysis, image enhancement, etc., can solve the problems of high crack identification misjudgment rate, inability to distinguish defects and workpiece contours well, and poor identification effect, etc. Achieve the effect of high recognition accuracy, fast calculation speed, and meet the accuracy requirements
Active Publication Date: 2020-12-29
SHANGHAI JIAO TONG UNIV +1
View PDF6 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
This scheme proposes a nonlinear fusion algorithm based on three morphological characteristic parameters, which has fast calculation speed and high stability. However, when the surface contour of the workpiece is complex and there are many interferences from pseudo cracks, the defect and the contour of the workpiece cannot be well distinguished. The effect is not good, and the misjudgment rate of crack identification is high
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0058] In this embodiment, a carbon steel workpiece is taken as an example to detect cracks on the surface of the carbon steel workpiece.
[0059] An intelligent identification method for magnetic particle flaw detection, such as figure 1 shown, including the following steps:
[0060] S1: Use the Bayesian optimization method to obtain the optimal parameter values of the ultraviolet light source light intensity, camera shooting angle and exposure time under the current magnetic particle concentration, including the following steps:
[0061] S101: Measure the current concentration of the magnetic powder liquid, and obtain the current fluorescent magnetic particle image by using the initial light intensity of the ultraviolet light source, camera shooting angle and exposure time, or manually select a set of parameter values for shooting.
[0062] S102: Compare the average grayscale of the current fluorescent magnetic particle image with the average grayscale of the image unde...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
| Property | Measurement | Unit |
|---|---|---|
| greyscale | aaaaa | aaaaa |
Login to View More
Abstract
The invention relates to an intelligent identification method for magnetic powder inspection. The method comprises the following steps: shooting a fluorescent magnetic powder image and an active whitelight image of a detected object; calculating the ratio of the change rate difference of the sub-channel gray values of the pixel points in the fluorescent magnetic powder image relative to the active white light image to obtain a binary image; carrying out binarized image denoising; obtaining a seed point image according to the gray value and the gray gradient of the pixel point; carrying out regional growth on the seed point image and then trimming; and extracting and predicting characteristic parameters of the suspected crack skeleton through a neural network model. Compared with the priorart, the suspected crack area is coarsely positioned through the ratio of the gray value change rate difference values of the sub-channels, the suspected crack skeleton is finely positioned through trimming after the seed point image area of the suspected crack area grows, full-image search is changed into region-of-interest search, and intrinsic characteristic parameters are designed for the suspected crack skeleton. Finally, the neural network is used for discrimination, cracks and pseudo cracks can be effectively distinguished, and the recognition precision is high.
Description
technical field [0001] The invention relates to the field of discriminating defects on the surface of metal workpieces, in particular to an intelligent identification method for magnetic particle flaw detection. Background technique [0002] There are various methods for workpiece crack detection. Compared with expensive ultrasonic flaw detectors, fluorescent magnetic particle crack detection is widely used due to its advantages of low cost, high sensitivity and fast detection speed. Fluorescent magnetic particle detection is mainly used in the detection of magnetic materials. After magnetizing ferromagnetic workpieces, pour fluorescent magnetic suspension liquid on the surface of the workpieces, and form a leakage magnetic field on the surface of the magnetized workpieces. Strong, so under the irradiation of black light, the defect magnetic marks appear yellow-green. [0003] The traditional fluorescent magnetic particle surface crack detection mainly relies on manual disc...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06T7/11G06T7/187G06N7/00G01N27/84
CPCG06T7/0004G06T7/11G06T7/187G01N27/84G06T2207/10004G06T2207/10024G06T2207/30164G06T2207/20076G06T2207/20081G06N7/01
Inventor 王福林蔡艳迟长云田华
Owner SHANGHAI JIAO TONG UNIV