红外焦平面阵列非均匀性多点实时压缩校正研究

非均匀性校正是提高红外图像质量的主要方法之一.针对传统的多点校正方法运算量大、实时性能差等缺点,提出多点实时压缩校正新方法,该方法将校正曲线相同或相似的不同探测单元归为一类,并将归类后的校正曲线的映射表存储在查找表中,校正过程中仅对查找表中的数据进行实时读取.实验结果表明:该算法计算量小、校正速度快,更加适合硬件实时实现.

Study on Real-time Multi-points Compressive Nonuniformity Correction of IRFPA

Nonuniformity correction is one of the main methods that can improve the quality of infrared image. Aiming at the problem of the large data storing and poor real-time capacity of traditional multi-point correction, a new multi-points compressive correction method is proposed. Detector pixels with identical or similar calibration curve will be classified as a class, and the mapping table of the categorized correction curves will be stored in a look-up table. During the correction process, only the data in the look-up table is read in real-time. The experimental results show the algorithm has less computation and requires less storage space of data, which is more suitable for real-time hardware implementation.