基于目标对象的遥感图像检索

针对有目标对象的遥感图像的特点,将图像分割成目标区域和背景区域,并在检索过程中赋予不同的权重,着重考虑目标对象的颜色、形状及空间信息特征,这样就可以避免当目标对象占整幅图像比重较小时,其特征信息容易被淹没在背景当中而造成的“伪误检”.实验结果表明,当图像中目标对象所占比重较小时,该方法与颜色直方图相比,能够更准确地查找出用户所需内容的图像.

Remote Sensing Image Retrieval Based on Target Object

The image is divided into target and background region aimed at remote sensing image with target object,and the pixels of the target and background are endowed with different powers during the retrieving.The feature of the target pixels is emphasized,and therefore,the error of conflicting between result of machine retrieval and human vision is avoided,while the proportion of the target is very small and the proportion of the background is much bigger in the whole image. The experimental results show that the method is more accurate and efficient in retrieving the interested images than color histogram method when there are small objects in the image.