CAJ | 학술논문

为了实现红外探测器的自适应校正，提出了一种基于小波变换的中值直方图规定非均匀性自适应校正方法。新算法利用正交小波变换分解图像，分别统计分解后图像像素的时域直方图，对像素邻域直方图进行排序得到中值，利用中值直方图灰度映射得到校正图像，最后把所有尺度校正后的图像反变换得到最终结果。实际应用证明，与同类自适应非均匀校正算法相比，该算法具有校正精度高、速度快的优点。
위료실현홍외탐측기적자괄응교정，제출료일충기우소파변환적중치직방도규정비균균성자괄응교정방법。신산법이용정교소파변환분해도상，분별통계분해후도상상소적시역직방도，대상소린역직방도진행배서득도중치，이용중치직방도회도영사득도교정도상，최후파소유척도교정후적도상반변환득도최종결과。실제응용증명，여동류자괄응비균균교정산법상비，해산법구유교정정도고、속도쾌적우점。
In order to realize infrared image adaptive correction, a method to adaptive correct the non-uniformity (NU) based on wavelet transform histogram normalization was introduced. The new algorithm used the orthogonal wavelet transform to analyze image, calculated the decomposition image histogram of the neighborhood elements of a single pixel, got median histogram by sorting pixels neighborhood histogram and completed decomposition image non-uniformity correction using the mean histogram normalization. Finally, a final result in all scales after the correction by using the inverse wavelet transform was obtained. According to the practical application, this algorithm has the advantage of fast convergence speed and high precision.