Infrared focal plane array heterogeneity self-adaptive correction method

Abstract

The present invention provides an infrared focal plane array nonuniform self-adapting correcting method pertaining to the field of infrared focal plane detector and the aim of the invention resides in controlling the degenerating of the object, improving the calculating speed and simplifying the hardware realization. The present invention includes an initialization step, a pre-processing step, a coarse correcting step and an exact correcting step. The method adopts an arithmetic combining one point correcting and a self-adapting correcting of neuronic network and containing an edge extracting module and a combined correcting arithmetic is obtained by combining the excellence of each arithmetic, thus, the space low-frequency noise is eliminated effectively and the object degenerating is controlled and the calculating speed is improved comparing to the neuronic network arithmetic directed by the edge.

Description

technical field [0001] The invention belongs to the field of infrared focal plane detectors, in particular to an adaptive correction method for non-uniformity of infrared focal plane arrays. Background technique [0002] Infrared imaging systems have evolved with the development of infrared detectors. In the first generation of infrared imaging systems, linear detectors are used to realize imaging through one-dimensional optical-mechanical scanning. With the maturity of CCD (Charge Coupled Device, electrically coupled device) related technologies, in the mid-1970s, the appearance of IRFPA (Infrared Focal Plane Array, infrared focal plane array) detector marked the second generation of infrared imaging system - staring The birth of the infrared imaging system. Compared with linear detectors, focal plane detector imaging has the advantages of high spatial resolution, strong detection capability, and high frame rate, and is rapidly becoming the mainstream device of infrared i...

AI Technical Summary

Problems solved by technology

[0004] The traditional neural network correction algorithm combines the characteristics of spatial domain processing (using the 4-neighborhood spatial mean as the expected output) and time domain processing (using neural network error feedback to realize the iterative update of the correction coefficient), and adaptively realizes non-uniformity correction. There are also some deficiencies: when obtaining the expected corrected output of the pixel (i, j), the spatial mean value of its four neighbors is directly used, which makes the algorithm present the characteristics of spatial low-pass filtering. When the scene in the sequence image is still When , there will be a degradation problem at the edge of the target

Under the premise of accurately obtaining the edge information of the scene, the algorithm can better suppress the degradation and artifacts of the target edge, and especially has a good effect on retaining weak and bright targets, but the calculation speed is relatively slow.

Images