论文部分内容阅读
受到多种因素的影响,单特征无法准确描述遥感图像信息,导致遥感图像的检索错误率高,为了提高遥感图像的检索正确率,提出一种基于特征聚合的遥感图像检索方法。首先分别提取遥感图像的颜色和纹理特征,然后通过欧式距离度量遥感图像之间的相似度,得到遥感图像的检索结果,最后采用神经网络对遥感图像检索结果进行反馈。仿真实验表明,本文方法可以准确找到用户需要的遥感图像,而且检索的时间比较短、检索效率高,具有很好的实用性。
Affected by many factors, single feature can not accurately describe remote sensing image information, resulting in high retrieval error rate of remote sensing images. In order to improve the retrieval accuracy of remote sensing images, a remote sensing image retrieval method based on feature aggregation is proposed. Firstly, the color and texture features of the remote sensing image are extracted respectively. Then the similarity between the remote sensing images is measured by the Euclidean distance to get the retrieval result of the remote sensing image. Finally, the neural network is used to feed back the retrieval result of the remote sensing image. The simulation results show that this method can find the remote sensing image that the user needs accurately, and the searching time is short and the retrieval efficiency is high, so it has good practicability.