新的Otsu阈值改进方法的红外小目标检测

针对红外小目标区域灰度对比度很差的现象,以及红外小目标实时性检测的要求,在对Otsu阈值方法进行研究的基础上,结合红外小目标图像的自身特点,提出了一种新的Otsu阈值改进的红外小目标检测方法。该方法不仅继承了Otsu阈值方法比较简单,计算速度较快的优点,更重要的是很好地解决了照度不均匀的图像分割时候多个红外小目标粘连的问题,使红外小目标能够被清晰地检测出来。实验结果表明,新的Otsu阈值改进方法检测出红外小目标的准确率比Otsu阈值检测方法提高了20%。

New Otsu threshold improvement method for detecting small infrared targets

In view of poor gray scale contrast of small IR targets and the requirement of real-time detection, combining with the behavior of small IR target image, a new Otsu threshold improvement method for detecting small IR targets is proposed. The method not only inherits the advantages of simplicity and fast calculation speed, but also solves the adjoining conglutination problems of multiple small IR targets when segmenting image with non-uniformity illumination, thus detecting small IR targets clearly. The experimental results show the accuracy of detecting small IR targets with the new Ostu threshold improvement method is 20% better than that of the old Otsu threshold detection method.