Industrial camera vignetting correction method

Abstract

The invention discloses an industrial camera vignetting correction method, which comprises the steps of 1, sequentially determining key rows and key columns of a calibration image according to a set minimum grid width and pixel gray values of pixel points in the calibration image, and forming grids according to the key rows and the key columns; 2, according to the preset neighborhood semi-axis length L and the row number and the column number of the calibration image, determining the neighborhood range of each grid point in the grid, calculating the neighborhood mean value of a correlation point group in the grid in the neighborhood range, wherein the correlation point group comprises four adjacent grid points; and 3, calculating a vignetting correction coefficient corresponding to each pixel point in the to-be-corrected image according to the neighborhood mean value, and performing vignetting correction on the to-be-corrected image according to the vignetting correction coefficient. According to the technical scheme of the invention, the proper correction coefficient is designed for the vignetting feature, so that the vignetting distortion of the image edge is corrected, the image quality is improved, the occupation of industrial camera hardware resources is reduced, and the image correction efficiency is improved.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular, to a method for correcting vignetting of industrial cameras. Background technique [0002] Automatic optical inspection is an effective inspection method for industrial automation. Machine vision using industrial cameras as a platform is used as a standard inspection technology. It is widely used in printing and packaging quality control, PCB inspection, rapid prototyping and other manufacturing fields. Taking printing and packaging quality control as an example, the system collects image information on the surface of the output printed matter through industrial cameras and lenses, and controls the quality of the printed matter through image processing such as positioning, identification, and classification. [0003] With the application of automatic optical inspection in a wider range of industries, the requirements for the quality of collected images are also g...

AI Technical Summary

Problems solved by technology

But there are also obvious deficiencies: 1. This correction method requires that the amount of correction coefficient data is consistent with the image resolution, and the amount of correction coefficient data is relatively large. At the same time, its data type is floating-point data, which increases the storage requirements of FPGA and other hardware.

2. This correction method requires FPGA to perform floating-point multiplication, which greatly occupies computing resources and affects the realization of other functions of the camera.

3. The multiplication coefficient of each pixel in this correction method does not take into account the noise in the space domain, and is extremely sensitive to camera bad points or small disturbances in the punctuation plane, thus affecting the quality of subsequent images

This method can correct the image vignetting to a certain extent, but there are several shortcomings: 1. The model used in this method needs to specify the image coordinates of the model center. When there is an error between the center of the vignetting feature and the center of the model, the calculated The correction coefficient does not match the actual situation, which reduces the stability and reliability of vignetting correction; 2. The model mentioned in this method adopts radial expansion, which cannot be reasonably corrected for non-centrosymmetric vignetting features, and the scope of application is limited; 3. The model analysis and calculation complexity of this method is high, and it is difficult to deploy in the camera hardware. The correction can only be completed on the computer terminal, and the offline deployment of the correction cannot be realized.

Images