A pcb image positioning correction method
A correction method and image positioning technology, applied in image enhancement, image analysis, image data processing, etc., can solve the problems of low accuracy, large amount of calculation, and strengthening the weight of weak edge components, so as to improve efficiency and accuracy, and accurately The effect of improving efficiency and reducing costs
Active Publication Date: 2021-02-23
FUJIAN UNIV OF TECH
View PDF8 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0013] The traditional Sobel operator is simple and easy to implement. Due to the average operation in the algorithm, it can smooth and denoise the image, but the Sobel operator cannot distinguish the foreground and background areas of the image well.
At the same time, since only the gradient information in two directions is considered, less consideration is given to the edge direction characteristics of the image, and some edge details are often lost. In a dynamic environment, the actual image gradient formula is unknown, so it is calculated according to the traditional Sobel operator The accuracy rate is not high; in addition, the calculation of the edge strength in formula (4) is calculated by the second-order norm, and involves the square root operation, and the calculation amount is large; at the same time, when calculating the edge strength, the edges in two directions are taken The weight coefficients of the components are the same, therefore, the weight of the strong edge components is weakened, and the weight of the weak edge components is strengthened
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
[0036] The process flow of the PCB image positioning correction method proposed by the present invention is as follows: figure 1 shown. The method includes three steps: image preprocessing, PCB area positioning, and PCB area-based image correction.
[0037] 1.1 Image preprocessing
[0038] The process of image preprocessing is as follows: figure 2 As shown, the input PCB image in this embodiment is an RGB image, and compared with other channels, the image details obtained by the G channel image are clearer. Therefore, in the image preprocessing step, the G channel image of the input RGB image is extracted to directly replace the common grayscale operation with the G channel image. In order to further improve the real-time performance of the algorithm, in the image preprocessing step, this embodiment also down-samples the extracted G-channel image, so as to reduce the number of pixels to be processed subsequently.
[0039] 1.2 PCB area positioning
[0040] PCB area positi...
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
Login to View More Abstract
The invention relates to the technical field of PCB defect detection, in particular to a PCB image positioning correction method. The PCB image positioning correction method includes an image preprocessing step, a PCB area positioning step, and a PCB area-based image correction step, and is characterized in that: in the PCB area positioning step, PCB area edges are detected based on Sobel operators in at least four directions . In the above technical solution, the operator template is first improved, and the original two-direction operator is improved to a multi-direction operator to overcome the traditional Sobel edge detection algorithm, which only considers the gradient information in two directions, and the accuracy is not high. Defects.
Description
technical field [0001] The invention relates to the technical field of PCB defect detection, in particular to a PCB image positioning correction method. Background technique [0002] At present, PCB positioning correction is required before PCB defect detection, and the correction effect directly affects the effect of later defect detection. The current traditional image positioning correction method has problems such as low accuracy rate, low automation level of mechanically fixed image correction method, and limited PCB board placement, resulting in low efficiency and accuracy of defect detection. [0003] PCB area positioning mainly includes edge extraction, image binarization and median filtering, and corresponding contour information is obtained. The Sobel operator in the edge detection algorithm has been widely used. For example, the application publication number CN 108985305A, the invention patent application published on December 11, 2018, discloses a laser-etched ...
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 Patents(China)
IPC IPC(8): G06T5/00G06T7/00G06T7/13
CPCG06T7/0004G06T7/13G06T2207/30141G06T5/80
Inventor 陈健朱寒
Owner FUJIAN UNIV OF TECH