论文部分内容阅读
针对传统的目标识别已不能适应信息化战争的需求,军事变形伪装目标的识别技术亟待加强等问题,对变形伪装军事目标的识别技术进行分析与研究,采用SIFT算子进行图像特征提取并匹配,采用快速排序法对识别目标按照相似度排序,并开发了一套军事变形伪装识别系统。通过系统测试可知,该系统能够有效地在海量图像库中识别出存在变形伪装的军事目标,并能够找出相似的军事目标,按照相似程度将这些目标排序,具有很好的应用价值。
In view of the traditional target recognition can not meet the needs of information war, the recognition technology of military deformation camouflage target needs to be strengthened, the discerning camouflage military target recognition technology is analyzed and studied, the SIFT operator is used to extract and match the image features, A rapid sequencing method was used to sort the recognition targets according to the similarity, and a military deformation camouflage recognition system was developed. The system test shows that the system can effectively identify the military targets with deformation camouflage in the mass image database, and can find similar military targets. According to the similar degree, these systems are of great value in application.