Measurement of Fluorescein Angiograms of the Optic Disc and Retina Using Computerized Image Analysis

Nagin, Paul; Schwartz, Bernard; Reynolds, George O.

doi:10.1016/s0161-6420(85)33999-4

Cited by 39 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Digital imaging of ocular images has become usual in ophthalmic practice in recent years (1,14,15,22). Efforts have been done to detect and quantify automatically several anatomical and pathological structures of ocular fundus [exudates (25), capillary network (6,7,19), drusen (9), macular ischemia (13), microaneurysms (23)].…”

Section: Discussionmentioning

confidence: 99%

Macular Edema Computer-Aided Evaluation in Ocular Vein Occlusions

Martínez-Costa

Marco

Ayala

et al. 1998

Computers and Biomedical Research

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Macular Edema Computer-Aided Evaluation in Ocular Vein Occlusions

Martínez-Costa

Marco

Ayala

et al. 1998

Computers and Biomedical Research

View full text Add to dashboard Cite

“…Of course, neither the determination of landmark points, nor the determination of individual defonnations are easy. One of the simplest solutions to both problems is the interactive approach, in which a human expert chooses corresponding landmark points in both images 11 .2l. In most registration applications, the subjectivity of a human expert (as well as time and economic considerations) is objectionable and a fully automated method is preferable.…”

Section: Local Methodsmentioning

confidence: 99%

“…Landmark based registration methods are commonly used for so-called "rubber-sheet" transformation?, although they have also been used for the determination of a single global registration for the whole image 11 .24. In our application, the objects of interest (patches of golden reflections) are relatively large (5 to 12 pixels wide) with no distinct boundaries, and the images are noisy, therefore making accurate automated (or manual) landmark-choosing very difficult Therefore.…”

Section: Local Methodsmentioning

confidence: 99%

<title>Registration of high-resolution images of the retina</title>

Cideciyan

Jacobson

Kemp

et al. 1992

Medical Imaging VI: Image Processing

View full text Add to dashboard Cite

A method of image registration is presented for the case when the deformation between two images can be well approximated with a combination of translation, rotation and global scaling. The method achieves very high accuracy by combining a global optimization in the 4-dimensional discrete parameter space with a local optimization in the 4-dimensional continuous parameter space. The 4-dimensional global optimization is accomplished with two 2-dimensional optimizations. The Fourier magnitude is used to decouple translation from rotation and scaling, and a log-polar mapping of the Fourier magnitude is used to convert rotation and scaling into shifts. Optimal rotation and scaling parameters are determined with a cross-correlation in the log-polar domain. After compensation for rotation and scaling differences, cross-correlation in the spatial domain yields the translation parameters. The four registration parameters are further refined with a local optimization using the correlation coefficient as a similarity measure in the 4-dimensional continuous parameter space. Results are shown from simulations and from registration of retinal images. For simulated images with a signal-to-noise ratio of -5 dB, the accuracy of the registration method is estimated 10 be better than 0.07 degrees, 0.1 %, and 0.3 pixels for rotation, scaling, and translation, respectively. In the case of 512x512 pixel images the computation resource requirements are compatible with high end PCs, i.e., approximately 25 minutes on an Intel 80486(33MHz based IBM/PC compatible.

show abstract

“…One class is based on feature extraction followed by registration using a transformation. For example, several authors [21,19,18,13,22,29,30,14,[3][4][5][6] considered extraction of the vessel structures followed by registration using a global transformation. Peli et al used vessel identification followed by feature based sequential detection [21].…”

Section: Introductionmentioning

confidence: 99%

“…Peli et al used vessel identification followed by feature based sequential detection [21]. Nagin and co-workers used edge enhancement and correlation to extract small sections of blood vessels, followed by a maximization of cross-correlation between small sections of the image [19]. Markow et al employed cross-correlation and edge detection to generate blood vessel templates in both reference and distorted images, followed by cross-correlation of these templates in the images to be registered [18].…”

Section: Introductionmentioning

confidence: 99%