Method, device, equipment and medium for real-time correction of linear array detector image mosaic

Abstract

The present application provides a real-time correction method, device, equipment and medium for image splicing of a line array detector, by obtaining the imaging data collected in real time by the line array detector; performing pre-correction processing on the imaging data; The pixel values ​​of the pixel point set to be corrected in each frame of the imaging data are corrected by gradient interpolation. This application can well avoid the influence of PD leakage current, electronic circuit noise, temperature change and PD module manufacturing process on the image splicing without increasing other costs such as hardware, and can not only ensure the correction speed well And calibration quality, but also to achieve good application results.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a real-time correction method, device, equipment and medium for image splicing of line array detectors. Background technique [0002] X-ray linear array detectors are commonly used in security inspection, vehicle inspection, food inspection and industrial non-destructive inspection and other fields. The X-ray line array detector imaging system is composed of a data acquisition card and several line array detector modules. A line array detector module is composed of multiple line array detector modules and circuit boards. Multiple line array detector modules The detector module is connected to the data acquisition card through a "daisy chain" interface, and the data acquisition card uploads the data of all linear array detector modules to the PC to form an image. Due to the daisy-chain working method of the line array detector modules and the cutting process at t...

AI Technical Summary

Problems solved by technology

Due to the daisy-chain working method of the line array detector modules and the cutting process at the edge of each line array detector module, the image output by the line array detector will always form a fixed double image where the line array detectors are spliced. Bad lines will lead to poor consistency between the dark field background data output by the linear array detector and the data after responding to X-rays, which will lead to poor image quality when detecting objects, and it is difficult to achieve true detection results and detection purposes

And because the pixel size of the line array detector is large, the application frame rate is high, and the degree of broken lines will increase with the increase of the detector's working temperature, these factors will make the image of the line array detector module stitching Calibration becomes more difficult

[0003] The traditional direct interpolation method (acquisition of the data of pixels around the splicing of the line array detector module in the same frame to interpolate and correct the double bad lines at the splicing), although the correction speed is very fast and the correction effect is very good in some cases, but when When an object enters the detection area obliquely, due to the large difference in gray values ​​on both sides of the object at the bad line, direct single-frame interpolation will cause gaps in the correction point, which will affect object recognition

Using multi-frame interpolation with threshold or secondary edge correction, there will be incomplete correction or unreasonable selection of threshold

Because linear array detectors are generally dual-energy detectors, high and low energies have different responses to X-rays, two sets of thresholds or coefficients may be required, and the difference in image gray value will increase with the increase of detector temperature and X-ray. Varies with light dose, these factors can make the threshold difficult to establish and not long-term accurate

Images