专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US08937849B1 Auto-focus for circular synthetic aperture sonar 有权
标题翻译：圆形合成孔径声纳自动对焦
公开(公告)号：US08937849B1
公开(公告)日：2015-01-20
申请号：US13568677
申请日：2012-08-07
申请人： Timothy M. Marston
发明人： Timothy M. Marston
IPC分类号： G01S15/89
CPC分类号： G01S7/52003 , G01S15/89
摘要： A method of focusing fully-coherent circular synthetic aperture sonar (CSAS) imagery is provided. A k-space representation of the CSAS image is generated using the two dimensional Fast-Fourier Transform (FFT). Sub-aperture images are generated by windowing the k-space representation and applying the two dimensional inverse FFT to the windowed spectrum. All adjacent complex sub-aperture images are correlated, the correlation peaks are detected and the relative shifts in X and Y as a function of look-angle are recorded. The relative shifts between adjacent sub-apertures are integrated and the means are subtracted to find the absolute shift as a function of look-angle. A motion-solution is calculated by exploiting the relationship between apparent scene shift and actual vehicle sway. The motion estimation is used to generate a phase-correction matrix that is multiplied by the k-space representation of the fully-coherent image.
摘要翻译：提供了一种聚焦全相干圆形合成孔径声纳（CSAS）图像的方法。使用二维快速傅里叶变换（FFT）生成CSAS图像的k空间表示。通过加窗k空间表示并将二维逆FFT应用于窗口频谱来生成子孔径图像。所有相邻的复合子孔径图像相关，记录相关峰值，记录作为视角的函数的X和Y的相对移动。相邻子孔之间的相对偏移被积分，并且减去平均值以找出作为视角的函数的绝对偏移。通过利用表观场景偏移与实际车辆摇摆之间的关系来计算运动解。运动估计用于产生乘以全相干图像的k空间表示的相位校正矩阵。

2. 发明授权

US10049295B2 Automated change detection for synthetic aperture sonar 有权
公开(公告)号：US10049295B2
公开(公告)日：2018-08-14
申请号：US15235251
申请日：2016-08-12
申请人： Tesfaye G-Michael , Daniel Sternlicht , Bradley Marchand , James Derek Tucker , Timothy M. Marston
发明人： Tesfaye G-Michael , Daniel Sternlicht , Bradley Marchand , James Derek Tucker , Timothy M. Marston
IPC分类号： G06K9/00 , G06K9/46 , G06T7/00
摘要： Methods and systems detect changes occurring over time between synthetic aperture sonar (SAS) images. A processor performs coarse navigational alignment, fine-scale co-registration and local co-registration between current image data and historical image data. Local co-registration includes obtaining correlation peaks for large neighborhood non-overlapping patches. Relative patch translations are estimated and parameterized into error vectors. Interpolation functions formed from the vectors re-map the current image onto the same grid as the historical image and the complex correlation coefficient between images is calculated. The resulting interferogram is decomposed into surge and sway functions used to define the argument of a phase function, which is multiplied by the current image to remove the effects of surge and sway on the interferogram. Based on the aforementioned computations, a canonical correlation analysis is performed to detect scene changes between the historical and new SAS images.

3. 发明申请

US20180046880A1 Automated Change Detection for Synthetic Aperture Sonar 审中-公开
公开(公告)号：US20180046880A1
公开(公告)日：2018-02-15
申请号：US15235251
申请日：2016-08-12
申请人： Tesfaye G-Michael , Daniel Sternlicht , Bradley Marchand , James Derek Tucker , Timothy M. Marston
发明人： Tesfaye G-Michael , Daniel Sternlicht , Bradley Marchand , James Derek Tucker , Timothy M. Marston
IPC分类号： G06K9/46 , G06K9/00 , G06T7/00
CPC分类号： G06K9/4671 , G06K9/00543 , G06T7/0057 , G06T7/32 , G06T7/33 , G06T7/521 , G06T2207/30212
摘要： Methods and systems detect changes occurring over time between synthetic aperture sonar (SAS) images. A processor performs coarse navigational alignment, fine-scale co-registration and local co-registration between current image data and historical image data. Local co-registration includes obtaining correlation peaks for large neighborhood non-overlapping patches. Relative patch translations are estimated and parameterized into error vectors. Interpolation functions formed from the vectors re-map the current image onto the same grid as the historical image and the complex correlation coefficient between images is calculated. The resulting interferogram is decomposed into surge and sway functions used to define the argument of a phase function, which is multiplied by the current image to remove the effects of surge and sway on the interferogram. Based on the aforementioned computations, a canonical correlation analysis is performed to detect scene changes between the historical and new SAS images.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式