欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
SIFT（scale invariant feature transform）算法因其有效的尺度、旋转、亮度、仿射、噪声等不变性,在模式识别和图像匹配领域中被广泛采用,但其实现过程需要在整个尺度空间上进行,时间复杂度相对较高,占用内存资源较大。针对SIFT算法对大幅面无人机航空遥感影像进行匹配时,在特征检测阶段容易产生内存溢出,导致无法进行匹配的问题,通过对大幅面影像进行分块,并考虑了处理块接边重叠问题,提出一种基于图像分块的Large-SIFT算法。实验表明,该算法在特征检测阶段不受内存限制,能较快速地对大幅面无人机航空遥感影像进行自动匹配,并在实际应用中为空中三角测量提供连接点数据。
英文摘要：
The SIFT operator is popularly used in pattern recognition and image matching since it is invariant to image rotation,scale,addition of noise,illumination changes and affine distortion. But the process requires to calculate the whole scale spaces of the input image that will cost very expensive time and take up a lot of memory resources. Especially for the large aerial image,the SIFT operator cannot process and doesn＇t generate keypoints. Therefore the paper implemented the improved SIFT operator based on image block for large-format aerial images： Large-SIFT. Experiments show that the algorithm can quickly and accurately detect keypoints in large-format aerial images for automatic matching. The results are the fundamental data for aerial triangulation in practical applications.
同期刊论文项目
同项目期刊论文
MNDISI: a multi-source composition index for impervious surface area estimation at the individual ci
BASI: An Index to Extract Built-Up Areas from High-Resolution Remote Sensing Images by Visual Salien
Sparse dimensionality reduction of hyperspectral image based on semi-supervised local Fisher discrim
BASI:An Index to Extract Built-Up Areas from High-Resolution Remote Sensing Images by Visual Salienc
期刊信息
位置：