遥感测绘图像特征智能识别技术

Intelligent Recognition Technology of Remote Sensing Mapping Image Features

为提高遥感测绘图像特征智能识别的平均识别率,研究一种遥感测绘图像特征智能识别技术。首先采用Malcov随机场模型描述纹理特征,选取有代表性的特征提取图像特征点,对特征向量进行降序,得到一个全新的特征向量集合,然后对特征向量进行归一化,提出一种基于大数据集域的快速自适应图像特征计算识别方法,实现智能系统对遥感测绘图像特征识别。结果显示此次研究的遥感测绘图像特征智能识别技术能够有效提高图像的平均识别率,并减少图像识别时间,具有一定应用价值。

In order to improve the average recognition rate of remote sensing mapping image feature intelligent recognition,a remote sensing mapping image feature intelligent recognition technology is designed.Firstly,the Malcov random field model is used to describe the texture features,the representative features are selected to extract the image feature points,and the feature vectors are in descending order to get a new feature vector set.Then,the feature vectors are normalized,and a fast adaptive image feature calculation and recognition method based on large data set is proposed to realize the feature recognition of remote sensing mapping image by intelligent system.The results show that the research of remote sensing mapping image feature intelligent recognition technology can effectively improve the average recognition rate of the image,and reduce the image recognition time,has a certain application value.