A method and device for infrared image processing

Abstract

The present invention provides an infrared image processing method and device. The method includes: acquiring an infrared image sequence including a current frame; segmenting the infrared image of the current frame according to the infrared image sequence to obtain its first original position area, Its first current location area and its first background area, wherein the first current location area is the area where the moving object in the infrared image is currently located, the first original location area is the moving object in the infrared image passing through other areas, and the first background area. The area is the area that the moving object does not pass through in the infrared image; the first filtering is performed on the first original position area of ​​the current frame by using the filtering kernel function and FFT, wherein, outside the set frequency domain range, the filtering kernel function is fast decline. The method and device of the present invention reduce noise in the infrared image, remove the ghost at the first original position of the moving object in the infrared image, and accurately acquire the current image of the moving object from the infrared image.

Description

technical field [0001] The present application relates to the field of image processing, and in particular, to a method and device for infrared image processing. Background technique [0002] Compared with visible light detectors, infrared detectors have a short development cycle, and still have problems such as high noise and insufficient resolution. Infrared detectors are thermal detection type detectors, which are greatly affected by temperature. The refrigerated infrared detector developed in the early stage has a refrigerator component, and the temperature of the detector can be stably cooled to about 80K, that is -193 ℃ when the device is working. Therefore, compared with the uncooled infrared detector, the cooled infrared detector It is less affected by temperature and therefore less noisy. But cooled infrared detectors are extremely expensive due to materials and refrigerators, so uncooled infrared detectors are being vigorously researched. Since there is no refri...

AI Technical Summary

Problems solved by technology

However, cooling infrared detectors are extremely expensive due to materials and refrigerators, so uncooled infrared detectors are being vigorously studied

Since there is no refrigerator structure, the uncooled infrared detector uses circuit design for thermal noise balance, and its ability to suppress noise is limited, and most of the noise is caused by the time-domain fringe noise introduced by temperature fluctuations

At present, the time domain denoising method will introduce ghosting and other problems, which will affect the recognition of the target. However, if the algorithm is not used for denoising, the noise will also seriously affect the picture quality; if the update iteration is performed on the hardware side, the time period will be long , the effect cannot be evaluated, etc.; therefore, there is an urgent need for a method to remove ghost images from the picture

Images