Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Intensity-Mosaic: automatic panorama mosaicking of disordered images with insufficient features

Intensity-Mosaic: automatic panorama mosaicking of disordered images with insufficient features Abstract.Purpose: Handling low-quality and few-feature medical images is a challenging task in automatic panorama mosaicking. Current mosaicking methods for disordered input images are based on feature point matching, whereas in this case intensity-based registration achieves better performance than feature-point registration methods. We propose a mosaicking method that enables the use of mutual information (MI) registration for mosaicking randomly ordered input images with insufficient features.Approach: Dimensionality reduction is used to map disordered input images into a low dimensional space. Based on the low dimensional representation, the image global correspondence can be recognized efficiently. For adjacent image pairs, we optimize the MI metric for registration. The panorama is then created after image blending. We demonstrate our method on relatively lower-cost handheld devices that acquire images from the retina in vivo, kidney ex vivo, and bladder phantom, all of which contain sparse features.Results: Our method is compared with three baselines: AutoStitch, “dimension reduction + SIFT,” and “MI-Only.” Our method compared to the first two feature-point based methods exhibits 1.25 (ex vivo microscope dataset) to two times (in vivo retina dataset) rate of mosaic completion, and MI-Only has the lowest complete rate among three datasets. When comparing the subsequent complete mosaics, our target registration errors can be 2.2 and 3.8 times reduced when using the microscopy and bladder phantom datasets.Conclusions: Using dimensional reduction increases the success rate of detecting adjacent images, which makes MI-based registration feasible and narrows the search range of MI optimization. To the best of our knowledge, this is the first mosaicking method that allows automatic stitching of disordered images with intensity-based alignment, which provides more robust and accurate results when there are insufficient features for classic mosaicking methods. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Medical Imaging SPIE

Intensity-Mosaic: automatic panorama mosaicking of disordered images with insufficient features

Download PDF
16 pages

Loading next page...
 
/lp/spie/intensity-mosaic-automatic-panorama-mosaicking-of-disordered-images-0QSwr5xyXa

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
SPIE
Copyright
© 2021 Society of Photo-Optical Instrumentation Engineers (SPIE)
ISSN
2329-4302
eISSN
2329-4310
DOI
10.1117/1.jmi.8.5.054002
Publisher site
See Article on Publisher Site

Abstract

Abstract.Purpose: Handling low-quality and few-feature medical images is a challenging task in automatic panorama mosaicking. Current mosaicking methods for disordered input images are based on feature point matching, whereas in this case intensity-based registration achieves better performance than feature-point registration methods. We propose a mosaicking method that enables the use of mutual information (MI) registration for mosaicking randomly ordered input images with insufficient features.Approach: Dimensionality reduction is used to map disordered input images into a low dimensional space. Based on the low dimensional representation, the image global correspondence can be recognized efficiently. For adjacent image pairs, we optimize the MI metric for registration. The panorama is then created after image blending. We demonstrate our method on relatively lower-cost handheld devices that acquire images from the retina in vivo, kidney ex vivo, and bladder phantom, all of which contain sparse features.Results: Our method is compared with three baselines: AutoStitch, “dimension reduction + SIFT,” and “MI-Only.” Our method compared to the first two feature-point based methods exhibits 1.25 (ex vivo microscope dataset) to two times (in vivo retina dataset) rate of mosaic completion, and MI-Only has the lowest complete rate among three datasets. When comparing the subsequent complete mosaics, our target registration errors can be 2.2 and 3.8 times reduced when using the microscopy and bladder phantom datasets.Conclusions: Using dimensional reduction increases the success rate of detecting adjacent images, which makes MI-based registration feasible and narrows the search range of MI optimization. To the best of our knowledge, this is the first mosaicking method that allows automatic stitching of disordered images with intensity-based alignment, which provides more robust and accurate results when there are insufficient features for classic mosaicking methods.

Journal

Journal of Medical Imaging – SPIE

Published: Sep 1, 2021

Keywords: automatic mosaicking; low-quality images; insufficient features; disordered input; mutual information

References

  • Automatic panoramic image stitching using invariant features
    Brown, M.; Lowe, D. G.
  • Stitching and surface reconstruction from endoscopic image sequences: a review of applications and methods
    Bergen, T.; Wittenberg, T.
  • In vivo micro-image mosaicing
    Loewke, K. E.
  • Image mosaicing of low quality neonatal retinal images
    Bontala, A.; Sivaswamy, J.; Pappuru, R. R.
  • Retinal image mosaicking using scale-invariant feature transformation feature descriptors and voronoi diagram
    Jalili, J.
  • Creating full view panoramic image mosaics and environment maps
    Szeliski, R.; Shum, H.-Y.
  • Real-time video mosaicking to guide handheld in vivo microscopy
    Yin, C.
  • Automated video-mosaicking approach for confocal microscopic imaging : an approach to address challenges in imaging living tissue and extend field of view
    Kose, K.
  • Real-time image mosaicing for medical applications
    Westwood, J.; Westwood, J. D.
  • Mosaic blood vessels in tumors: frequency of cancer cells in contact with flowing blood
    Chang, Y. S.
  • Image alignment and stitching: a tutorial
    Szeliski, R.
  • Graphcut textures: image and video synthesis using graph cuts
    Kwatra, V.
  • Seamless image stitching of scenes with large motions and exposure differences
    Eden, A.; Uyttendaele, M.; Szeliski, R.
  • Poisson image editing
    Pérez, P.; Gangnet, M.; Blake, A.
  • Cross correlation versus mutual information for image mosaicing
    Ghannam, S.; Abbott, A. L.
  • Underwater video mosaics as visual navigation maps
    Gracias, N.; Santos-Victor, J.
  • Video mosaicing using a mutual information-based motion estimation process
    Dame, A.; Marchand, E.
  • Mosaicing of bladder endoscopic image sequences: distortion calibration and registration algorithm
    Miranda-Luna, R.
  • Second-order optimization of mutual information for real-time image registration
    Dame, A.; Marchand, E.
  • Fast construction of panoramic images for cystoscopic exploration
    Hernandez-Mier, Y.
  • Comparative study of image registration techniques for bladder video-endoscopy
    Hamadou, A. B.
  • Mosaicing of medical video-endoscopic images: data quality improvement and algorithm testing
    Miranda-Luna, R.
  • Automatic image alignment and stitching of medical images with seam blending
    Kumar, A.
  • Retinamatch: efficient template matching of retina images for teleophthalmology
    Gong, C.
  • Analysis of a complex of statistical variables into principal components
    Hotelling, H.
  • Kernel principal component analysis
    Schölkopf, B.; Smola, A.; Müller, K.-R.
  • A global geometric framework for nonlinear dimensionality reduction
    Tenenbaum, J. B.; De Silva, V.; Langford, J. C.
  • Randomized algorithms for the low-rank approximation of matrices
    Liberty, E.
  • Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
    Halko, N.; Martinsson, P.-G.; Tropp, J. A.
  • Randomized nonnegative matrix factorization
    Erichson, N. B.
  • Spline pyramids for intermodal image registration using mutual information
    Thevenaz, P.; Unser, M. A.
  • Predicting error in rigid-body point-based registration
    Fitzpatrick, J. M.; West, J. B.; Maurer, C. R.
  • Structured analysis of the retina
    Goldbaum, M.
  • Robust principal component analysis?
    Candès, E. J.
  • Feasibility of a video-mosaicking approach to extend the field-of-view for reflectance confocal microscopy in the oral cavity in vivo
    Peterson, G.
  • Handheld reflectance confocal microscopy for the detection of recurrent extramammary paget disease
    Yélamos, O.
  • Peri-operative delineation of non-melanoma skin cancer margins in vivo with handheld reflectance confocal microscopy and video-mosaicking
    Flores, E.
  • Handheld line-scanned dual-axis confocal microscope with pistoned mems actuation for flat-field fluorescence imaging
    Wei, L.
  • Cystoscopy
    Engelsgjerd, J.; Deibert, C.
  • Cancer statistics, 2021.
    Siegel, R. L.
  • Recurrence of high-risk bladder cancer: a population-based analysis
    Chamie, K.
  • Bladder cancer (Version 6.2020)
  • Bladder Cancer, Version 3.2020, NCCN Clinical Practice Guidelines in Oncology
    Flaig, T. W.
  • 3D reconstruction of cystoscopy videos for comprehensive bladder records
    Lurie, K. L.
  • Chapter 4–endoscopic treatment of bladder tumors
    Geavlete, P. A.
  • Autostitch
    Brown, M.