Academia.eduAcademia.edu

Outline

Title
Abstract
Key Takeaways
Figures
References

Super-Resolution Image Reconstruction: A Technical Overview

Profile image of Yana AlshevskayaYana Alshevskaya

Abstract
sparkles

AI

This paper provides a technical overview of super-resolution (SR) image reconstruction techniques, highlighting the need for high-resolution (HR) images across various applications, such as medical imaging and computer vision. It discusses limitations associated with traditional methods of increasing image resolution, including sensor constraints and cost factors, and emphasizes that SR aims to reconstruct HR images from multiple low-resolution (LR) images through advanced signal processing methods.

Key takeaways
sparkles

AI

  1. Super-resolution (SR) techniques enhance spatial resolution using multiple low-resolution images captured from the same scene.
  2. Optimal pixel size limitation is around 40 µm² for CMOS processes, restricting sensor manufacturing methods.
  3. Signal processing methods like SR image reconstruction offer cost-effective solutions for improving image quality without new hardware.
  4. Accurate motion estimation between low-resolution images is crucial for successful SR image reconstruction.
  5. Current challenges in SR focus on noise modeling, registration errors, and efficiency in computation.
Figures (13)
A 1. Basic premise for super resolution.
A 1. Basic premise for super resolution.
A 3. Observation model relating LR images to HR images. Consider the desired HR image of size L, N, x L, N, written in lexicographical notation as the vector X=[%, ,X 55Xy |’, where N=L,N, xL, N,.Namely, x is the ideal undegraded image that is sampled at or above the Nyquist rate from a continuous scene which is assumed to be bandlimited. Now, let the parameters L, and L, represent the down-sampling factors in the obser- vation model for the horizontal and vertical directions, re- Let us consider the system matrix involved in (1). The motion that occurs during the image acquisition 1s repre- sented by warp matrix M, . It may contain global or local translation, rotation, and so on. Since this information is
A 3. Observation model relating LR images to HR images. Consider the desired HR image of size L, N, x L, N, written in lexicographical notation as the vector X=[%, ,X 55Xy |’, where N=L,N, xL, N,.Namely, x is the ideal undegraded image that is sampled at or above the Nyquist rate from a continuous scene which is assumed to be bandlimited. Now, let the parameters L, and L, represent the down-sampling factors in the obser- vation model for the horizontal and vertical directions, re- Let us consider the system matrix involved in (1). The motion that occurs during the image acquisition 1s repre- sented by warp matrix M, . It may contain global or local translation, rotation, and so on. Since this information is
A 5. Low-resolution sensor PSF. A 4. The necessity of interpolation in HR sensor grid.
A 5. Low-resolution sensor PSF. A 4. The necessity of interpolation in HR sensor grid.
generally unknown, we need to estimate the scene mo- tion for each frame with reference to one particular frame. The warping process performed on HR image x is actu- ally defined in terms of LR pixel spacing when we esti- mate it. Thus, this step requires interpolation when the fractional unit of motion is not equal to the HR sensor grid. An example for global translation is shown in Figure 4. Here, a circle (O) represents the original (reference) FIR image x, and a triangle (A) and a diamond (©) are globally shifted versions of x. If the down-sampling factor is two, a diamond (©) has (0.5, 0.5) subpixel shift for the horizontal and vertical directions and a triangle (A) has a shift which is less than (0.5,0.5). As shown in Figure 4, a diamond (©) does not need interpolation, but a triangle (A) should be interpolated from x since it is not located on the HR grid. Although one could use ideal interpola- tion theoretically, in practice, sumple methods such as
generally unknown, we need to estimate the scene mo- tion for each frame with reference to one particular frame. The warping process performed on HR image x is actu- ally defined in terms of LR pixel spacing when we esti- mate it. Thus, this step requires interpolation when the fractional unit of motion is not equal to the HR sensor grid. An example for global translation is shown in Figure 4. Here, a circle (O) represents the original (reference) FIR image x, and a triangle (A) and a diamond (©) are globally shifted versions of x. If the down-sampling factor is two, a diamond (©) has (0.5, 0.5) subpixel shift for the horizontal and vertical directions and a triangle (A) has a shift which is less than (0.5,0.5). As shown in Figure 4, a diamond (©) does not need interpolation, but a triangle (A) should be interpolated from x since it is not located on the HR grid. Although one could use ideal interpola- tion theoretically, in practice, sumple methods such as
A 7. Registration-interpolation-based reconstruction. A 6. Scheme for super resolution.
A 7. Registration-interpolation-based reconstruction. A 6. Scheme for super resolution.
Ur and Gross [5] performed a nonuniform interpola- tion of an ensemble of spatially shifted LR images by uti- lizing the generalized multichannel sampling theorem of Papoulis [73] and Brown [74]. The interpolation is fol- lowed by a deblurring process, and the relative shifts are assumed to be known precisely here. Komatsu et al. [1] presented a scheme to acquire an improved resolution im- age by applying the Landweber algorithm [75] from multiple images taken simultaneously with multiple cam- eras. They employ the block-matching technique to mea- sure relative shifts. If the cameras have the same aperture, however, it imposes severe limitations both in their ar- rangement and in the configuration of the scene. This dif- ficulty was overcome by using multiple cameras with different apertures [6]. Hardie et al. developed a tech- nique for real-time infrared image registration and SR re- construction [7]. They utilized a gradient-based
Ur and Gross [5] performed a nonuniform interpola- tion of an ensemble of spatially shifted LR images by uti- lizing the generalized multichannel sampling theorem of Papoulis [73] and Brown [74]. The interpolation is fol- lowed by a deblurring process, and the relative shifts are assumed to be known precisely here. Komatsu et al. [1] presented a scheme to acquire an improved resolution im- age by applying the Landweber algorithm [75] from multiple images taken simultaneously with multiple cam- eras. They employ the block-matching technique to mea- sure relative shifts. If the cameras have the same aperture, however, it imposes severe limitations both in their ar- rangement and in the configuration of the scene. This dif- ficulty was overcome by using multiple cameras with different apertures [6]. Hardie et al. developed a tech- nique for real-time infrared image registration and SR re- construction [7]. They utilized a gradient-based
A 9. Aliasing relationship between LR image and HR image.
A 9. Aliasing relationship between LR image and HR image.
registration algorithm for the acquired frames and neighbor interpolation ap estimating the shifts between presented a weighted nearest proach. Finally, Wiener filter- ing is applied to reduce effects of blurring and noise caused by the system. Shah and Zakhor proposed an SR color video enhancement a gorithm using the Landweber algorithm [8]. They also consider the inaccuracy of the registration algorithm by finding a set of candidate mo-
registration algorithm for the acquired frames and neighbor interpolation ap estimating the shifts between presented a weighted nearest proach. Finally, Wiener filter- ing is applied to reduce effects of blurring and noise caused by the system. Shah and Zakhor proposed an SR color video enhancement a gorithm using the Landweber algorithm [8]. They also consider the inaccuracy of the registration algorithm by finding a set of candidate mo-
A 10. Regularized SR reconstruction results by (a) nearest neigh- bor interpolation, (b) CLS with small regularization parameter, (©) CLS with large regularization parameter, and (d) MAP with edge-preserving prior. = Bayesian estimation methods are used when the a pos- teriori probability density function (PDF) of the original
A 10. Regularized SR reconstruction results by (a) nearest neigh- bor interpolation, (b) CLS with small regularization parameter, (©) CLS with large regularization parameter, and (d) MAP with edge-preserving prior. = Bayesian estimation methods are used when the a pos- teriori probability density function (PDF) of the original
&\ 11. POCS SR results (a) by bilinear interpolation and by POCS after (b) 10 iterations, (c) 30 iterations, and (d) 50 iterations.
&\ 11. POCS SR results (a) by bilinear interpolation and by POCS after (b) 10 iterations, (c) 30 iterations, and (d) 50 iterations.
A 12. A pictorial example of IBP method.
A 12. A pictorial example of IBP method.
\ 13. The effect of registration error in HR estimate; result by (a) the conventional algorithm and (b) the channel-adaptive reg- ularization.
\ 13. The effect of registration error in HR estimate; result by (a) the conventional algorithm and (b) the channel-adaptive reg- ularization.

Related papers

Process and pixels for high performance imager in SOI-CMOS technology
Bedabrata Pain

2003

We present the pixel and process design and test-results from imager pixels implemented in SOI-CMOS technology. While bulk-CMOS foundries have made various alterations in their process flow to integrate a CMOS imager frontend with high performance processors, and spectacular progress has been made in making large format imagers, several major challenges remain. These pertain to obtaining low and anomaly-free dark-current, low cross-talk, high linearity, good response uniformity and high dynamic range. Non-planarity in the junction area, resulting from LOCOS or STI isolation, high doping, and high electric fields cause problems for dark current reduction, while simultaneously obtaining small depletion widths for deep sub-micron FETs and large depletion widths around the photodiode junction cause problems for cross-talk and quantum efficiency.

View PDFchevron_right
A mega-pixel high speed CMOS imager with sustainable Gigapixel/sec readout rate
Peter Seitz, Stefan Lauxtermann

A high-speed active pixel (APS) imager with a physical resolution of 1024x768 pixels is presented which acquires up to 1000 frames/sec. The imager has a physical resolution of 1024x768 pixels and acquires up to 1000 frames/sec. The light sensitive photodiodes in the array are arranged in a chessboard like pattern and are color masked with an RGB Bayer pattern. Interpolation then leads to a resolution of 1536x1024 pixels. Analog data are transmitted off-chip through 32 output channels that operate at over 30 MHz each. For electronic exposure control the sensor can be operated with a rolling line as well as a global shutter. Also, the photo response is programmable to have a linear, logarithmic or combined lin-log characteristic. For processing a standard 0.5µm CMOS technology was chosen. The chip has physical dimensions of 14 x 20 mm 2 and dissipates 600 mW at full resolution and maximum speed. The sensor resolution can be reduced without having to modify the analog output rate, as l...

View PDFchevron_right
Limits in point to point resolution of MOS based pixels detector arrays
Nicolas Fourches

2017

In high energy physics point-to-point resolution is a key prerequisite for particle detector pixel arrays. Current and future experiments require the development of inner-detectors able to resolve the tracks of particles down to the micron range. Present-day technologies, although not fully implemented in actual detectors, can reach a 5-μm limit, this limit being based on statistical measurements, with a pixel-pitch in the 10 μm range. This paper is devoted to the evaluation of the building blocks for use in pixel arrays enabling accurate tracking of charged particles. Basing us on simulations we will make here a quantitative evaluation of the physical and technological limits in pixel size. Attempts to design small pixels based on SOI technology will be briefly recalled here. A design based on CMOS compatible technologies that allow a reduction of the pixel size below the micrometer is introduced here. Its physical principle relies on a buried carrier-localizing collecting gate. Th...

View PDFchevron_right
Interlacing in charge-coupled imaging devices
Carlo H. SEQUIN

IEEE Transactions on Electron Devices, 2000

AbsirucbSolid-state imaging devices to be used in commercial broadcast TV or the Picturephone" system have to supply the video information in a 2:1 interlaced timing format. An efficient way to achieve such a readout from charge-coupled area imaging devices of the frame transfer type is described. The resolution cells of the device are elongated in the vertical dimension to extend across the distance corresponding to two scanning lines in the display. For the two fields forming a full frame the charge generated by the incident light is integrated alternately underneath different sets of electrodes.

View PDFchevron_right
Super-resolution imaging technique based on a LCoS display: Increase of CCD resolution limit
Mohammad sohail

The resolution of any imaging system is limited by many factors, as by the diffraction resolution limit of the system or by resolution limitations related to the geometry of the optical elements. One of the main geometrical limitations in optical systems is due to CCD cameras used for the image acquisition. In particular, pixel size, shape and pixel pitch of the used CCD camera always impose a limit in resolution. This paper faces the problem of the geometric super-resolution limitation, providing an approach free of mechanical movements, which helps to overcome the problems related to CCD pixels pitch. To this aim, a parallel aligned (PA) liquid crystal on silicon (LCoS) display is placed at the Fourier plane of a transparent object, and different linear phases are addressed to it. Afterwards, an image-forming optical system provides the final image of the object at the CCD camera plane. By addressing different linear phases to the LCoS display, object images with different sub-pixel displacements in 1-D are acquired by the CCD camera. Afterwards, all the shifted images are combined, leading to a final super-resolved image with larger dimension than the original object image. In addition, an inverse filtering process is also included into the proposed method, leading to certain extent, to a decrease of the blurring effect. The experimental comparison of the object images obtained with and without using the proposed technique provides the improvement, in terms of resolution, achieved by applying our technique.

View PDFchevron_right
A single-chip programmable digital CMOS imager with enhanced low-light detection capability
Julie Heynssens

VLSI Design 2000. Wireless and Digital Imaging in the Millennium. Proceedings of 13th International Conference on VLSI Design

View PDFchevron_right
Limits in point to point resolution of MOS (Metal-Oxide-Semiconductor) based pixels detector arrays
Nicolas Fourches

2018

In high energy physics point to point resolution is a key prerequisite for particle detector pixel arrays. Current and future experiments require the development of inner-detectors able to resolve the tracks of particles down to the micron range. Present-day technologies, although not fully implemented in actual detectors can reach a 5 micometer limit, based on statistical measurements, with a pixel-pitch in the 10 micrometer range. Attempts to design small pixels based on SOI (Silicon On Insulator) technology will be briefly recalled here. This paper is devoted to the evaluation of the building blocks with regard to their use in pixel arrays for the accurate tracking of the charged particles. We will make here a simulations based quantitative evaluation of the physical limits in the pixel size. A design based on CMOS (Complementary Metal-Oxide-Semiconductor) compatible technologies that allows a reduction of the pixel size down to the sub-micronmeter range is introduced. Its physic...

View PDFchevron_right
High-resolution CMOS imaging detector
Gene P Weckler

Proceedings of SPIE - The International Society for Optical Engineering

A 2.5 cm by 5 cm, 512 by 1024 pixel CMOS photodiode array detector designed specifically for digital radiography will be described. All necessary scanning and readout circuitry is integrated within the detector. The small pixel spacing of 48 micrometers allows the imager to easily achieve the 10 lp/mm resolution required for the targeted interventional mammography application. Direct coupling to the scintillator and a pixel fill factor of more than 80% leads to high DQE over a large range of exposure values. The detector exhibits very low dark current of about 30 pA/cm2 at room temperature, which allows for low-noise operation and long integration times. Read noise of less than 200 electrons rms and a saturation level of 2.8 X 106 electrons combine for a large dynamic range greater than 80 dB. The conversion gain of the detector is 0.5 (mu) V/electron. Combined with a Gd2O2S scintillator, the imager achieves up to 25% MTF at 10 lp/mm and a DC detective quantum efficiency of 50%. The...

View PDFchevron_right
<title>Progress in CMOS active pixel image sensors</title>
Bedabrata Pain

Proceedings of SPIE, 1994

View PDFchevron_right
Enhanced resolution for future fabrication
Chang-Lee Chen

IEEE Circuits and Devices Magazine, 2003

he introduction a few years ago of 180 nm design rule CMOS technology, using 248 nm stepper tools, marked an important milestone in optical lithography. For the first time, devices were being fabricated commercially with feature sizes smaller than the exposure wavelength. To achieve the required resolution and process control using standard methods is difficult, so the use of optical resolution-enhancement methods, such as optical proximity correction, off-axis illumination, and phase-shift masks has been increasing [1]. These techniques have enabled the current production of 130 nm design rule CMOS using high numerical aperture (NA) 248 nm optical tools. Feature sizes of less than 50% of the exposure wavelength are currently being imaged in production using optical resolution-enhancement methods. A challenge of advanced device research has been the fabrication technology for the deeply scaled feature sizes required. Our research efforts as part of the DARPA Advanced Microelectronics (AME) program have necessitated gate resolutions down to 25 nm. Most fabrication approaches in this regime suffer from either low throughput (e-beam) or poor process latitude (feature "slimming" methods). Since our goal was

View PDFchevron_right

References (79)

  1. T. Komatsu, K. Aizawa, T. Igarashi, and T. Saito, "Signal-processing based method for acquiring very high resolution image with multiple cameras and its theoretical analysis," Proc. Inst. Elec. Eng., vol. 140, no. 1, pt. I, pp. 19-25, Feb. 1993.
  2. S. Borman and R.L. Stevenson, "Spatial resolution enhancement of low-resolution image sequences. A comprehensive review with directions for future research," Lab. Image and Signal Analysis, University of Notre Dame, Tech. Rep., 1998.
  3. S. Borman and R.L. Stevenson, "Super-resolution from image se- quences-A review," in Proc. 1998 Midwest Symp. Circuits and Systems, 1999, pp. 374-378.
  4. S. Chaudhuri, Ed., Super-Resolution Imaging. Norwell, MA: Kluwer, 2001.
  5. H. Ur and D. Gross, "Improved resolution from sub-pixel shifted pictures," CVGIP: Graphical Models and Image Processing, vol. 54, pp. 181-186, Mar. 1992.
  6. T. Komatsu, T. Igarashi, K. Aizawa, and T. Saito, "Very high resolution imaging scheme with multiple different-aperture cameras," Sinal Processing: Image Commun., vol. 5, pp. 511-526, Dec. 1993.
  7. M.S. Alam, J.G. Bognar, R.C. Hardie, and B.J. Yasuda, "Infrared image registration and high-resolution reconstruction using multiple translationally shifted aliased video frames," IEEE Trans. Instrum. Meas., vol. 49, pp. 915-923, Oct. 2000.
  8. N.R. Shah and A. Zakhor, "Resolution enhancement of color video se- quences," IEEE Trans. Image Processing, vol. 8, pp. 879-885, June 1999.
  9. N. Nguyen and P. Milanfar "An efficient wavelet-based algorithm for im- age superresolution," in Proc. Int. Conf. Image Processing, vol. 2, 2000, pp. 351-354.
  10. R.Y. Tsai and T.S. Huang, "Multipleframe image restoration and registra- tion," in Advances in Computer Vision and Image Processing. Greenwich, CT: JAI Press Inc., 1984, pp. 317-339.
  11. S.P. Kim, N.K. Bose, and H.M. Valenzuela, "Recursive reconstruction of high resolution image from noisy undersampled multiframes," IEEE Trans. Acoust., Speech, Signal Processing, vol. 38, pp. 1013-1027, June 1990.
  12. S.P. Kim and W.Y. Su, "Recursive high-resolution reconstruction of blurred multiframe images," IEEE Trans. Image Processing, vol. 2, pp. 534-539, Oct. 1993.
  13. N.K. Bose, H.C. Kim, and H.M. Valenzuela, "Recursive implementation of total least squares algorithm for image reconstruction from noisy, undersampled multiframes," in Proc. IEEE Conf. Acoustics, Speech and Signal Processing, Minneapolis, MN, Apr. 1993, vol. 5, pp. 269-272.
  14. S.H. Rhee and M.G. Kang, "Discrete cosine transform based regularized high-resolution image reconstruction algorithm," Opt. Eng., vol. 38, no. 8, pp. 1348-1356, Aug. 1999.
  15. M.C. Hong, M.G. Kang, and A.K. Katsaggelos, "A regularized multi- channel restoration approach for globally optimal high resolution video se- quence," in SPIE VCIP, vol. 3024, San Jose, CA, Feb. 1997, pp. 1306-1317.
  16. M.C. Hong, M.G. Kang, and A.K. Katsaggelos, "An iterative weighted regularized algorithm for improving the resolution of video sequences," in Proc. Int. Conf. Image Processing, vol. 2, 1997, pp. 474-477.
  17. M.G. Kang, "Generalized multichannel image deconvolution approach and its applications," Opt. Eng., vol. 37, no. 11, pp. 2953-2964, Nov. 1998.
  18. R.C. Hardie, K.J. Barnard, J.G. Bognar, E.E. Armstrong, and E.A. Wat- son, "High-resolution image reconstruction from a sequence of rotated and translated frames and its application to an infrared imaging system," Opt. Eng., vol. 37, no. 1, pp. 247-260, Jan. 1998.
  19. N.K. Bose, S. Lertrattanapanich, and J. Koo, "Advances in superresolution using L-curve," in Proc. Int. Symp. Circuits and Systems, vol. 2, 2001, pp. 433-436.
  20. B.C. Tom and A.K. Katsaggelos, "Reconstruction of a high-resolution image by simultaneous registration, restoration, and interpolation of low-resolution images," Proc. 1995 IEEE Int. Conf. Image Processing, vol. 2, Washington, DC, Oct. 1995, pp. 539-542.
  21. R.R. Schulz and R.L. Stevenson, "Extraction of high-resolution frames from video sequences," IEEE Trans. Image Processing, vol. 5, pp. 996-1011, June 1996.
  22. R.C. Hardie, K.J. Barnard, and E.E. Armstrong, "Joint MAP registration and high-resolution image estimation using a sequence of undersampled images," IEEE Trans. Image Processing., vol. 6, pp. 1621-1633, Dec. 1997.
  23. P. Cheeseman, B. Kanefsky, R. Kraft, J. Stutz, and R. Hanson, "Super-re- solved surface reconstruction from multiple images," NASA Ames Research Center, Moffett Field, CA, Tech. Rep. FIA-94-12, Dec. 1994.
  24. H. Stark and P. Oskoui, "High resolution image recovery from im- age-plane arrays, using convex projections," J. Opt. Soc. Am. A, vol. 6, pp. 1715-1726, 1989.
  25. A.M. Tekalp, M.K. Ozkan, and M.I. Sezan, "High-resolution image re- construction from lower-resolution image sequences and space varying im- age restoration," in Proc. IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP), San Francisco, CA., vol. 3, Mar. 1992, pp. 169-172.
  26. A.J. Patti, M.I. Sezan, and A.M. Tekalp, "Superresolution video recon- struction with arbitrary sampling lattices and nonzero aperture time," IEEE Trans. Image Processing, vol. 6, no. 8, pp. 1064-1076, Aug. 1997.
  27. P.E. Eren, M.I. Sezan, and A.M. Tekalp, "Robust, object-based high-resolution image reconstruction from low-resolution video," IEEE Trans. Image Processing, vol. 6, no. 10, pp. 1446-1451, Oct. 1997.
  28. A.J. Patti and Y. Altunbasak, "Artifact reduction for set theoretic super resolution image reconstruction with edge adaptive constraints and higher-order interpolants," IEEE Trans. Image Processing, vol. 10, no. 1, pp. 179-186, Jan. 2001.
  29. B.C. Tom and A.K. Katsaggelos "An iterative algorithm for improving the resolution of video sequences," in Proc. 1996 SPIE Conf. Visual Commu- nications and Image Processing, Orlando, FL, Mar. 1996, pp. 1430-1438.
  30. M. Elad and A. Feuer, "Restoration of a single superresolution image from several blurred, noisy, and undersampled measured images," IEEE Trans. Image Processing, vol. 6 , no. 12, pp. 1646-1658, Dec. 1997.
  31. M. Irani and S. Peleg, "Improving resolution by image registration," CVGIP: Graphical Models and Image Proc., vol. 53, pp. 231-239, May 1991.
  32. S. Mann and R.W. Picard, "Virtual bellows: Constructing high quality stills from video," in Proc. IEEE Int. Conf. Image Processing, Austin, TX, Nov. 1994, pp. 13-16.
  33. M. Irani and S. Peleg, "Motion analysis for image enhancement resolu- tion, occlusion, and transparency," J. Visual Commun. Image Represent., vol. 4, pp. 324-335, Dec. 1993.
  34. M. Elad and A. Feuer, "Superresolution restoration of an image sequence: adaptive filtering approach," IEEE Trans. Image Processing, vol. 8, pp. 387-395, Mar. 1999.
  35. M. Elad and A. Feuer, "Super-resolution reconstruction of image se- quences," IEEE Trans. Pattern Anal. Machine Intelli., vol. 21, no. 9, pp. 817-834, Sept. 1999.
  36. M.C. Chiang and T.E. Boult, "Efficient super-resolution via image warp- ing," Image Vision Computing, vol. 18, pp. 761-771, Dec. 2000.
  37. D. Rajan and S. Chaudhuri, "Generation of super-resolution images form blurred observations using an MRF model," J. Math. Imaging Vision, vol. 16, pp. 5-15, 2002.
  38. D. Rajan and S. Chaudhuri, "Simultaneous estimation of super-resolved intensity and depth maps from low resolution defocused observations of a scene," in Proc. IEEE int. Conf. Computer Vision, Vancouver, Canada, July 2001, pp. 113-118.
  39. D. Rajan and S. Chaudhuri, "Generalized interpolation and its applica- tions in super-resolution imaging," Image and Vision Computing, vol. 19, pp. 957-969, Nov. 2001.
  40. M.V. Joshi and S. Chaudhuri, "Super-resolution imaging: Use of zoom as a cue," in Proc. Indian Conf. Vision, Graphics and Image Processing, Ahmedabad, India, Dec. 2002, pp. 439-444.
  41. N.K. Bose, H.C. Kim, and B. Zhou, "Performance analysis of the TLS al- gorithm for image reconstruction from a sequence of undersampled noisy and blurred frames," in Proc. ICIP-94, IEEE Int. Conf. Image Processing, vol. 3, 1994, pp. 571-575.
  42. M. Ng, J. Koo, and N. Bose, "Constrained total least squares computa- tions for high resolution image reconstruction with multisensors," Int. J. Imaging Syst. Technol., vol. 12, pp. 35-42, 2002.
  43. M.K. Ng and N.K. Bose, "Analysis of displacement errors in high-resolu- tion image reconstruction with multisensors," IEEE Trans. Circuits Syst. I, vol. 49, pp. 806-813, June 2002.
  44. M. Park, E. Lee, J. Park, M.G. Kang, and J. Kim, "DCT-based high-resolution image reconstruction considering the inaccurate sub-pixel motion information," SPIE Optical Engineering, vol. 41, no. 2, pp. 370-380, Feb. 2002.
  45. W. Lim, M. Park, and M.G. Kang, "Spatially adaptive regularized iterative high resolution image reconstruction algorithm," in Proc. VCIP2001, Photonicswest, San Jose, CA, Jan. 2001, pp. 20-26.
  46. E. Lee and M.G. Kang, "Regularized adaptive high-resolution image re- construction considering inaccurate subpixel registration," IEEE Trans. Im- age Processing, to be published.
  47. Wirawan, P. Duhamel, and H. Maitre, "Multi-channel high resolution blind image restoration," in Proc. IEEE ICASSP, AZ, Nov. 1989, pp. 3229-3232.
  48. N. Nguyen, P. Milanfar, and G. Golub, "Efficient generalized cross-validation with applications to parametric image restoration and reso- lution enhancement," IEEE Trans. Image Processing, vol. 10, pp. 1299-1308, Sept. 2001.
  49. N. Nguyen, P. Milanfar, and G. Golub, "A computationally efficient superresolution image reconstruction algorithm," IEEE Trans. Image Pro- cessing, vol. 10, pp. 573-583, Apr. 2001.
  50. M. Elad and Y. Hel-Or, "A fast super-resolution reconstruction algorithm for pure translational motion and common space-invariant blur," IEEE Trans. Image Processing, vol. 10, no. 8, pp. 1187-1193, Aug. 2001.
  51. M. Ng, R. Chan, T. Chan, and A.Yip, "Cosine transform preconditioners for high resolution image reconstruction," Linear Algebra Appls., vol. 316, pp. 89-104, 2000.
  52. D.S. Messing and M.I. Sezan, "Improved multi-image resolution enhance- ment for colour images captured by single-CCD cameras," in Proc. Int. Conf. Image Processing, vol. 3, 2000, pp. 484-487.
  53. M.K. Ng, "An efficient parallel algorithm for high resolution color image reconstruction," in Proc. 7th Int. Conf. Parallel and Distributed Systems: Workshops, 2000, pp. 547-552.
  54. B.C. Tom and A.K. Katsaggelos, "Resolution enhancement of mono- chrome and color video using motion compensation," IEEE Trans. Image Processing, vol. 10, no. 2, pp. 278-287, Feb. 2001.
  55. M. Ng and W. Kwan, "High-resolution color image reconstruction with Neumann boundary conditions," Ann. Oper. Res., vol. 103, pp. 99-113, 2001.
  56. D. Chen and R.R. Schultz, "Extraction of high-resolution video stills from MPEG image sequences," in Proc. 1998 IEEE Int. Conf. Image Process- ing, vol. 2, 1998, pp. 465-469.
  57. Y. Altunbasak and A.J. Patti, "A maximum a posteriori estimator for high resolution video reconstruction from MPEG video," in Proc. 2000 IEEE Int. Conf. Image Processing, vol. 2, 2000, pp. 649-652.
  58. B. Martins and S. Forchhammer, "A unified approach to restoration, deinterlacing and resolution enhancement in decoding MPEG-2 video," IEEE Trans. Circuits Syst. Video Technol., vol. 12, no. 9, pp. 803-811, Sept. 2002.
  59. C.A. Segall, R. Molina, A.K. Katsaggelos, and J. Mateos, "Reconstruction of high-resolution image frames from a sequence of low-resolution and compressed observations," in Proc. 2002 IEEE Int. Conf. Acoustics, Speech, and Signal Processing, vol. 2, 2002, pp. 1701-1704.
  60. S.C. Park, M.G. Kang, C.A. Segall, and A.K. Katsaggelos, "Spatially adaptive high-resolution image reconstruction of low resolution DCT-based compressed images," in Proc. 2002 IEEE Int. Conf. Image Pro- cessing, vol. 2, 2002, pp. 861-864.
  61. B.K. Gunturk, Y. Altunbasak, and R.M. Mersereau, "Multiframe resolu- tion-enhancement methods for compressed video," IEEE Signal Processing Lett., vol. 9, pp. 170-174, June 2002.
  62. H.C. Andrews and B.R. Hunt, Digital Image Restoration. Englewood Cliffs, NJ: Prentice-Hall, 1977.
  63. A.K. Katsaggelos, Ed. Digital Image Restoration. Heidelberg, Germany: Springer-Verlag. Springer. vol. 23, 1991.
  64. I.J. Schoenberg, "Cardinal interpolation and spline functions," J. Approx. Theory., vol. 2, pp. 167-206, 1969.
  65. R.E. Crochiere and L.R. Rabiner, "Interpolation and decimation of digi- tal signals-A turorial review," Proc. IEEE, vol. 69, no. 3, pp. 300-331, Mar. 1981.
  66. M. Unser, A. Aldroubi, and M. Eden, "Enlargement or reduction of digi- tal images with minimum loss of information," IEEE Trans. Image Pro- cessing, vol. 4, no. 3, pp. 247-258, Mar. 1995.
  67. V.N. Dvorchenko, "Bounds on (deterministic)correlation functions with applications to registration." IEEE Trans. Pattern Anal. Machine Intell., vol. 5, no. 2. pp. 206-213, 1983.
  68. Q. Tian and M.N. Huhns, "Algorithm for subpixel registration," Com- puter Vision, Graphics, Image Proc., vol. 35, pp. 220-233, 1986.
  69. C.A. Bernstein, L.N. Kanal, D. Lavin, and E.C. Olson, "A geometric ap- proach to subpixel registration accuracy," Computer Vsion, Graphics, and Im- age Proc., vol. 40, pp. 334-360, 1987.
  70. L.G. Brown, "A survey of image registration techniques," ACM Comput. Surveys, vol. 24, no. 4, pp. 325-376, Dec. 1992.
  71. J. J. Clark, M R. Palmer, and P.D. Laurence, "A transformation method for the reconstruction of functions from nonuniformly spaced samples," IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-33, pp. 1151-1165, 1985.
  72. S.P. Kim and N.K. Bose, "Reconstruction of 2-D bandlimited discrete signals from nonuniform samples," Proc. Inst. Elec. Eng., vol. 137, pt. F, pp. 197-204, June 1990.
  73. A. Papoulis, "Generalized sampling theorem," IEEE Trans. Circuits Syst. vol. 24, pp. 652-654, Nov. 1977.
  74. J.L. Brown, "Multi-channel sampling of low pass signals," IEEE Trans. Circuits Syst., vol. CAS-28, pp. 101-106, Feb. 1981.
  75. L. Landweber, "An iteration formula for Fredholm integral equations of the first kind," Amer. J. Math. vol. 73, pp. 615-624, 1951.
  76. A.M. Tekalp, Digital Video Processing. Englewood Cliffs, NJ: Prentice Hall, 1995.
  77. P.C. Hansen, and D. Prost O'Leary, "The use of the L-curve in the regu- larization of discrete ill-posed problems," SIAM J. Sci. Comput., vol. 14, no. 6, pp. 1487-1503, Nov. 1993.
  78. H.J. Trussell and M.T. Civanlar, "Feasible solution in signal restoration," IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-32, pp. 201-212, Mar. 1984.
  79. M.G. Kang and A.K. Katsaggelos, "General choice of the regularization functional in regularized image restoration," IEEE Trans. Image Processing, vol. 4, pp. 594-602, May 1995.

Related papers

Increase of Charge-Coupled Device Resolution limits using Spatial Light Modulator
Mohammad Sohail

2016

Resolution of the imaging devices is mainly restricted by the pixels pitch, the pixels size and shape. We show a methodology where a Liquid Crystal on Silicon presentation is utilized to produce diverse direct stages at the article phantom plane, prompting distinctive sub-pixel relocations of the item picture at the picture plane. Along these lines, pictures of the same item with various movements are inspected by Charge-Coupled Device (CCD) camera. At long last, all the acquired pictures are legitimately consolidated prompting a last super-determined picture. Correlation of unique article and the one acquired with our methodology plainly affirm a noteworthy change in determination.

View PDFchevron_right
Geometric and Radiometric Evaluation of the Potential of a High Resolution Cmos -Camera
Farhad Samadzadegan

2010

The progress of digital imaging systems has created manifold opportunities and new applications to close range photogrammetry. The classical techniques have found significant changes and this seems to be an ongoing process which is heavily influenced by new technologies showing up on the sensor market. CMOS (Complimentary Metal Oxide Semiconductor) sensors are on of the challenges to CCD sensors as they are providing a similar performance at a cheaper price level. One of the demands is to investigate the radiometric and geometric characteristics of the imaging sensor. This includes considering other factors involved in image acquisition with CMOS sensors and requires a detailed analysis on the factors and its effects for the overall performance of CMOS cameras. This paper analyses the radiometric and geometric performance of the Canon EOS-1Ds, a high resolution CMOS camera. Presented are issues of restoration and geometric correction as well as an experimental investigation using di...

View PDFchevron_right
52 Megapixel APS-H-size CMOS Image Sensor for Super High Resolution Image Capturing
Masaaki Iwane

2007

We have developed a new CMOS image sensor having pixels of more than 52M in APS-H size. The CMOS image sensor has the most number of pixels known to date without stitching. The sensitivity of the monochromatic image sensor is 39000e/lx・s. The sensitivity of the color image sensor (green pixel) is 16600e/lx・s. Pixel size is 3.2um x 3.2um. Random noise is 5.5e with a saturation level of 24000e. The CMOS image sensor has 5 x 5 random block readout mode and 4(2×2) adjacent pixels averaging mode. The reproduced image shows splendid high resolution.

View PDFchevron_right
Charge-coupled imaging devices: Design considerations
M. Tompsett

IEEE Transactions on Electron Devices, 2000

In this paper some of the parameters relevant to the design of charge-coupled imaging devices are considered. Among these are charge storage capability, transfer efficiency, charge conservation, dark current, and the anticipated signal-to-noise ratio. Each is discussed, and the resultant effects on the performance of imaging devices are investigated.

View PDFchevron_right
Charge-coupled imaging devices: Experimental results
M. Tompsett

IEEE Transactions on Electron Devices, 2000

View PDFchevron_right
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved