J. Le Moigne. Parallel registration of multi-sensor remotely sensed imagery using wavelet coefficients. In Proceedings of the SPIE - The International Society for Optical Engineering, volume 2242, pages 432--43, Orlando, FL, USA, 5-8, April 1994.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....sensing satellite systems This work is supported by NASA and USRA CESDIS and the amount of images to be registered, the time required to register input images to existing reference images is quite large. This prompted the need for parallel processing to conquer the lengthy computation time[1][2][3] In this paper, we discuss the computational savings in image registration resulted from using the Wavelet based technique which exploits the multiresolution property of Wavelet [4] 5] 6] This iterative refinement Wavelet based method avoids exhaustive search for the relative orientation of ....

J. LeMoigne, "Parallel Registration of Multi-Sensor Remotely Sensed Imagery Using Wavelet Coefficients," in Proceedings 1994.

....representation of image data. Using such multi resolution data, the size of the search data can be reduced by initially searching at lowest resolution images (smallest data size) and then proceeding to higher resolution images where the search results are only refined [Akansu95] Corvi95] [LeMoigne94] [LeMoigne96] LeMoigne97] Wavelet based multi resolution preserves most of the important features of the original data even at a low resolution. It also eliminates weak features in higher resolution while highlighting strong image features [Mallat89] LeMoigne et al. presented a cross comparison ....

....to implement the parallel GA based 29 image registration. However, only one medical image data was used in their experiment. Based on the multi resolution wavelet (MRW) technique, LeMoigne presented a finegrain parallel algorithm for the MasPar SIMD (Single Program Multiple Data) architecture [LeMoigne94][LeMoigne95] Substantial computational savings have been reported. Unfortunately, the fine grain MRW is not applicable to modern MIMD architectures. A hardware based parallel image registration algorithm was proposed by Turton [Turton94] It was based on the GA based image registration presented ....

[Article contains additional citation context not shown here]

J. LeMoigne, "Parallel Registration of Multi Sensor Remotely Sensed Imagery Using Wavelet Coefficients," in Proceedings

.... This work is supported by NASA and USRA CESDIS Super Computing 1997 2 and the amount of images to be registered, the time required to register input images to existing reference images is quite large. This prompted the need for parallel processing to conquer the lengthy computation time[1][2][3] In this paper, we discuss the computational savings in image registration resulted from using the Wavelet based technique which exploits the multiresolution property of Wavelet [4] 5] 6] This iterative refinement Wavelet based method avoids exhaustive search for the relative orientation of ....

J. LeMoigne, "Parallel Registration of Multi-Sensor Remotely Sensed Imagery Using Wavelet Coefficients," in Proceedings

....1. Our image registration techniques utilize the multi resolution property in a wavelet domain and Genetic Algorithms (GAs) to reduce the search data size as well as the search scope. The compressed low resolution wavelet images are used for initial and intermediate iterative correlation [4] 5] 6][7][8] Starting from the lowest resolution level, the search results get refined as the search succeeds to the higher resolution wavelet images. This is called multiresolution iterative refinement algorithm (IRA) 7] 8] For subpixel accuracy in image registration, IRA has one drawback. The initial ....

....wavelet images are used for initial and intermediate iterative correlation [4] 5] 6] 7] 8] Starting from the lowest resolution level, the search results get refined as the search succeeds to the higher resolution wavelet images. This is called multiresolution iterative refinement algorithm (IRA) [7][8] For subpixel accuracy in image registration, IRA has one drawback. The initial search points are quite large and IRA covers almost all possible solutions in that level. Reducing the search points can lead to wrong solutions (local optima problem) Instead of linearly search, Genetic ....

[Article contains additional citation context not shown here]

J. LeMoigne, "Parallel Registration of Multi Sensor Remotely Sensed Imagery Using Wavelet Coefficients," in Proceedings

.... mother wavelet, and then by analyzing the transformed signals. In the image processing domain, Wavelet transforms have been proven to be very useful for such tasks as image compression and reconstruction, feature extraction, and image registration ( Chu92] Dja92] Dau92] Mal89] Str89] [Lem94]) Furthermore, the multi resolution scheme developed by Mallat ( Mal89] Cod92] Num92] provides a very fast algorithm which increases the importance of wavelets for on line processing of imagery data. The speed of such processing is especially important for managing remotely sensed data ....

J. LeMoigne, Parallel Registration of Multi-Sensor Remotely Sensed Imagery Using Wavelet Coefficients, Proceedings SPIE Wavelets'94, Orlando, April 5-8, 1994.

....of tuberculosis CAT or MRI or sonogram interpretation, etc. aerial or satellite photography Note the need to correct for distortion (due to height and angle of view as well as nonlinear distortion due to lens aberration or surface curvature) note also the registration problem (cf. e.g. [13]) in aligning adjacent mission storage of the image, we may expect an application map A : f = image 7 a = action : F A (1.3) where the space A = AA of possible actions may be either discrete (e.g. a recognition problem) or continuous (e.g. robotic control) The appropriate cost J 2 ....

J. LeMoigne, Parallel registration of multi-sensor remotely sensed imagery using wavelet coefficients, CESDIS report, NASA Goddard, 1994.

....to the tremendous amount of data generated from the EOS remote sensing satellites and the amount of images to be registered, the time required to register input images to existing reference images is quite large. This prompted the need for parallel processing to conquer the lengthy computation time[1][2] 3] 4] Genetic Algorithms (GAs) have been long known to be very robust for search and optimization problems. It is based on the principles of natural biological evolution that operates on a population of potential solutions applying the principle of survival of the fittest to produce better ....

J. LeMoigne, "Parallel Registration of Multi Sensor Remotely Sensed Imagery Using Wavelet Coefficients," in Proceedings

....amount of data generated from the MTPE remote sensing satellite systems and the amount of images to be registered, the time required to register input images to existing reference images is quite large. This prompted the need for parallel processing to conquer the lengthy computation time[1][2][3] In this paper, we discuss the computational savings in image registration resulted from using the Wavelet based technique which exploits the multiresolution property of Wavelet [4] 5] 6] This iterative refinement Wavelet based method avoids exhaustive search for the relative orientation of ....

J. LeMoigne, "Parallel Registration of Multi-Sensor Remotely Sensed Imagery Using Wavelet Coefficients," in Proceedings 1994 SPIE Wavelet Applications Conf., Orlando, pp. 432-443, 1994.

No context found.

J. Le Moigne. Parallel registration of multi-sensor remotely sensed imagery using wavelet coefficients. In Proceedings of the SPIE - The International Society for Optical Engineering, volume 2242, pages 432--43, Orlando, FL, USA, 5-8, April 1994.

No context found.

J. Le Moigne. Parallel registration of multi-sensor remotely sensed imagery using wavelet coefficients. In Proceedings of the SPIE - The International Society for Optical Engineering, volume 2242, pages 432--43, Orlando, FL, USA, 5-8, April 1994.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC