摘要
从自动目标识别原理出发,通过对变背景下斑点状散布目标特性的分析,发现了该类目标图像中背景、噪声和斑点状散布目标的固有规律。进而,从图像像素灰度这一最基本的要素着手,研究了如何在大量的序列图像中搜索目标帧位置的方法。然后,基于目标轮廓提取和串行边界分割技术,解决了变背景下斑点状散布目标自动识别的难题。实验结果证明:该算法具有高效、快速、稳定、适用等优点,并为相关领域内的应用提供了宝贵的参考价值。
Based on the principle of automatic target recognition and location, characters of the fleck discrete targets under changing backgrounds have been analyzed thoroughly with the discovery of the instinct features of backgrounds, noises and fleck discrete targets. Then the location of target's frame is solved by the pixel gray statistical changes in the vast image sequence. Thus determines fleck discrete targets' precision dimensions by feature extraction and serial borderline segmentation technologies. After critical experiments, it proves that this algorithm is fast, stable and efficient that provides possible further usage in other related fields.
出处
《航空计算技术》
2004年第3期57-60,共4页
Aeronautical Computing Technique