Course 0510.7211 “Digital Image Processing: Applications”. Lecture I. .... France.
Astronomers were among the first to employ the new imaging techniques. In.
L. Yaroslavsky
DIGITAL IMAGE PROCESSING: APPLICATIONS LECTURE NOTES
Course 0510.7211, Semester B
Tel Aviv University Faculty of Engineering, Department of Interdisciplinary Studies
L. Yaroslavsky. Course 0510.7211 “Digital Image Processing: Applications”
Lecture I. Images and Imaging Devices Images are signals produced by special devices, imaging devices Designing and perfecting imaging devices were among the main tasks of the modern science beginning from Galileo Galilei and Newton
Brief history of imaging devices • Eye-glasses(1-st century or earlier)
Pliny the Elder wrote in 23-79 A.D.: "Emeralds are usually concave so that they may concentrate the visual rays. The Emperor Nero used to watch in an Emerald the gladiatorial combats." The modern reinvention of spectacles occurred around 1280-1285 in Florence, Italy. It's uncertain who the inventor was. Some give credit to a nobleman named Amati (Salvino degli Armati, 1299 ). It has been said that he made the invention, but told only a few of his closest friends • Camera-obscura (pinhole camera) (Ibn Al Haytam, X century):
• Printing devices (Gutenberg). Painting art. Origin of the theory of imaging (Renaissance painters: Leonardo da Vinci, Albrecht Dürer, 15-16th century)
A woodcut by Albrecht Dürer showing “Perspektoraph” , an istrument used by painters in 16 century. Dürer was held a while for the inventing this instrument, it goes however back to Leonardo da Vinchi
•
Magnifying glass(around 1600 year), Microscope (Joannes and Zacharios Jansen, 1595; A. Leeuwenhoek, R. Hook, around 1660)
Inventor of optical microscope is not known. Credit for the first microscope is usually given to Dutch (from other sources, Middleburg, Holland) spectacle-maker Joannes and his son Zacharios Jansen. While experimenting with several lenses in a tube, they discovered (around the year 1595) that nearby objects appeared greatly enlarged. (partly adopted from [1]) . That was the forerunner of the compound microscope and of the telescope. The father of microscopy, Anthony Leeuwenhoek of Holland (1632-1723), started as an apprentice in a dry goods store where magnifying glasses were used to count the threads in cloth. He taught himself new methods for grinding and polishing tiny lenses of great curvature which gave magnifications up to 270, the finest known at that time. These led to the building of his microscopes and the biological discoveries for which he is famous. He was the first to see and describe bacteria, yeast plants, the teeming life in a drop of water, and the circulation Microscope of Hooke (R. Hooke, Micrographia, 1665) of blood corpuscles in capillaries. Robert Hooke, the English father of microscopy, re-confirmed Anthony van Leeuwenhoek's discoveries of the existence of tiny living organisms in a drop of water. Hooke made a copy of Leeunwenhoek's microscope and then improved upon his design
Modern Zeiss microscope
•
Telescope (Galileo, 1609 year)
Newton’s telescope-refractor
In 1609, Galileo, father of modern physics and astronomy, heard of these early experiments, worked out the principles of lenses, and made a much better instrument with a focusing device. The scientific impetus produced by the great discoveries made with the telescope can be gauged from the enthusiastic manner in which Huygens in the “Dioptrica” speaks of these discoveries. He describes how Galileo was able to see the mountains and valleys of the moon, to observe sunspots and determine the rotation of the sun, to discover Jupiter’s satellites and the phases of Venus, to resolve the Milky Way into stars, and to establish the differences in apparent diameter of the planets and fixed stars (after E. Mach, The principles of Physical Optics, Dover Publ., 1926).
Huygens (“Dioptrica, de telescopiis”) held the view that Only a superhuman genius could have invented the telescope on the basis of theoretical considerations, but the frequent use of spectacles and lenses of various shapes over a period of 300 years contributed to its chance invention.
Hubble space telescope: in principle, the same optics as in Newton’s telescope
Photography
(Niepps, 1826; Daguerre, 1836 year, W. F. Talbot, 1844.
First public report was presented by F. Arago, 19.8.1839 at a meeting of L’Institut, Paris, France) In the 19-th century, scientists began to explore ways of “fixing” the image thrown by a glass lens. (H. Nieps, 1826; J. Daguerr, 1836; W. F. Talbot, 1844) The first method of light writing was developed by the French commercial artist Louis Jacque Mande Daguerre (1787-1851). The daguerrotype was made on a shhet of silver-plated coper, which could be inked and then printed to produce accurate reproduction of original works or scenes. The surface of the copper was polished to a mirrorlike brilliance, then rendered light sensitive by treament with iodine fumes. The copper plate was then exposed to an image sharply focused by the camera’s well-ground, optically correct lens. The plate was removed from the camera and treated with mercury vapors to develop the latent image. Finally, the image was fixed by removal of the remaining photosensitive salts in a bath of hyposulfite and toned with gold chloride to improve contrast and durability. Color, made of powdered pigment, was applied directly to the metal surface with a finely pointed brush. Daguerre’s attempt to sell his process (the daguerreotype) through licensing was not successful, but he found an enthusiastic supporter in Francois Arago, an eminent member of the Academie des Sciences in France. Arago recommended that the French government compensate Daguerre for his considerable efforts, so that the daguerreotype process could be placed at the service of the entire world. The French government complied, and the process was widely publicized by F. Arago, 19.8.1839 at a meeting of L’Institut, Paris on August 19, 1839, as a gift to the world from France. Astronomers were among the first to employ the new imaging techniques. In 1839-1840, John W. Draper, professor of chemistry at New York University, made first photographs the moon in first application of daguerreotypes in astronomy. The photoheliograph, a device for taking telescopic photographs of the sun, was unveiled in 1854. In 1840 optical means used to reduce daguerreotype exposure times to 3-5 min. In 1841 William Henry Fox Talbot patents a new process involving creation of paper negatives. By the end of 19-th century, photography had become an important means for scientific research and also a commercial item that entered people every day life. It has been keeping this status till very recently. Invention of photography (combination of imaging optics + photo sensitive material) was a revolutionary step. Image formation and image display were separated. Photographic plate/film combines three basic imaging functions: image recording, image storage and image display
•
X-ray imaging (Wilhelm Conrad Röntgen, Nov. 8,1895; Institute of Physics, University of Würzburg, Germany, the 1-st Nobel Prize, 1901)
A new type of radiation for imaging was discovered X-ray point source+ photographic film or photo-luminescent screen
Wilhelm Conrad Röntgen
Fluorography 1907
One of the first medical X-ray images (a hand with small shots in it)
Fluorography 2000
Marie Curie, the discoverer of radium, operated and taught operating first X-ray imaging machines in French army during the 1-st world war
Photography had played a decisive role in the discovery of X-Rays. It had played a decisive role in yet another revolutionary discovery, the discovery of radioactivity. In 1896, Antoine Henri Becquerel accidentally discovered radioactivity while investigating phosphorescence in uranium salts. This discovery eventually led, along with other, to new imaging techniques,
radiography
Antoine Henri Becquerel Modern gamma-camera: Gamma-ray collimator + Gamma-ray-to light converter + photo sensitive array + CRT as a display. Collimator separates rays from different object points.
• Electronic imaging Electron microscope (Ernst Ruska, 1931, The Nobel Prize, 1986) Electron optics + luminescent screen or electron sensitive array + CRT display
Transmission Electron Microscope: Atoms of gold (Au_clusters) on MoS2.
Scanning electron microscope image (from http://www.sst.ic.ac.uk/intro/AFM.htm )
Electronic television Video camera: imaging optics +electron optics+scanning+ photo-electronic converter ; image display – CRT tube. An important step: image discretization. ~1910, Boris Lvovich Rosing, St. Petersburg, Russia: Cathode Ray Tube as a display device ~1920, His former student, US émigré, Vladimir Kozmich Zvorykin – conversion of optical image into electric signal and inverse conversion: iconoscope& kinescope. David Sarnov, sun of a rabbi from Belorussia, US émigré and former telegraph operator, President of RCA at that time, invited Zvorykin to RCA and gave him $100.000 for the development of a commercial electronic television system ~1935 : first regular TV broadcasting in Britain, Germany, USA ~ end of 1940-th – commercial TV broadcasting ~ end of 1950-ths – colour television
One of the first TV image (V.K. Zvorykin, 1933)
• Radars (around 1935-40), Sonars. The principle of active vision was invented and implemented: Radio wave beam forming antenna + space scanning mechanism + receiver + CRT as a display
•
Acoustic microscope (1950-th, after R. Bracewell, Two-dimensional Imaging, Prentice Hall, 1995): Electric Oscillator
Sapphir (Al2O3) rode
Receiver
Piezo-electric transducer (niobium i )
Movable specimen (immersed in a li id)
A monochromatic sound pulse can be focused to a point on the solid surface of an object by a lens (sapphire rode), and the reflection will return to the lens to be gathered by a receiver. The strength of the reflection depends on the acoustical impedance looking into the solid surface relative to the impedance of the propagating media. If the focal point performs a raster scan over the object, a picture of the surface impedance is formed. Acoustic impedance of a medium depends on its density and elastic rigidity. Acoustic energy that is not reflected at the surface but enters the solid may be only lightly attenuated and then reflect from surface discontinuities to reveal an image of the invisible interior. With such a device, an optical resolution can be achieved. A major application is in the semiconductor industry for inspecting integrated circuits. The idea of focusing an acoustic beam was originally suggested by Rayleigh. The application of scanning acoustic microscopes goes back to 1950. A scanning optical microscope can also be made on the same principle. It has value as a means of imaging an extended field without aberrations associated with a lens.
• Scanned-proximity probe (SPP) microscopes.
SPP- microscopes work by measuring a local property - such as height, optical absorption, or magnetism - with a probe or "tip" placed very close to the sample. The small probe-sample separation (on the order of the instrument's resolution) makes it possible to take measurements over a small area. To acquire an image the microscope raster-scans the probe over the sample while measuring the local property in question. Scanned-probe systems do not use lenses, so the size of the probe rather than diffraction effects generally limit their resolution
Tunnel microscope (1980-th; The Nobel prize 1986)
Schematic of the physical principle and initial technical realization of Scanning Tunnel Microscope. (a) shows apex of the tip (left) and the sample surface (right) at a magnification of about 108. The solid circles indicate atoms, the dotted lines electron density contours. The path of the tunnel current is given by the arrow. (b) Scaled down by factor of 104. The tip (left) appears to touch the surface (right). (c) STM with rectangular piezo drive X,Y,Z of the tunnel tip at left and “loose” L (electrostatic “motor”) for rough positioning (µm to cm range) of the sample S (from G. Binning, H. Rohrer: Physica 127B, 37, 1984)
Scanning tunnel microscope image of silicon surface. The image shows two single layer steps (the jagged interfaces) separating three terraces. Because of the tetrahedral bonding configuration in the silicon lattice, dimer tow directions are orthogonal on terraces joined by a single layer step. The area pictured is 30x30 nm
A conductive sample and a sharp metal tip, which acts as a local probe, are brought within a distance of a few ångstroms, resulting in a significant overlap of the electronic wave functions (see the figure). With applied bias voltage (typically between 1mV and 4V), a tunelling current (typically between 0.1nA and 10 nA) can flow from the occupied electronic states near the Fermi level of one electrode into the unoccupied states of the other electrode. By using a piezo-electric drive system of the tip and a feedback loop, a map of the surface topography can be obtained. The exponential dependence of the tunneling current on the tipto-sample spacing has proven to be the key for the high spatial resolution which can be achieved with the STM. Under favorable conditions, a vertical resolution of hundredths of an ångstrom and the lateral resolution of about one ångstrom can be reached. Therefore, STM can provide real-space images of surfaces of conducting materials down to the atomic scale. (from R. Wiesendanger and H.-J. Güntherodt, Introduction, Scanniing Tunneling Microscopy I, General Principles and Applications to Clean and Absorbate-Covered Surfaces, Springer Verlag, Berlin, 1994)
Atomic force microscope
(after http://stm2.nrl.navy.mil/how-afm/how-afm.html).
The atomic force microscope is one of about two dozen types of scanning probe microscopes. AFM operates by measuring attractive or repulsive forces between a tip and the sample. In its repulsive "contact" mode, the instrument lightly touches a tip at the end of a leaf spring or "cantilever" to the sample. As a raster-scan drags the tip over the sample, some sort of detection apparatus measures the vertical deflection of the cantilever, which indicates the local sample height. Thus, in contact mode the AFM measures hard-sphere repulsion forces between the tip and sample. In noncontact mode, the AFM derives topographic images from measurements of attractive forces; the tip does not touch the sample. AFMs can achieve a resolution of 10 pm, and unlike electron microscopes, can image samples in air and under liquids. To achieve this most AFMs today use the optical lever. The optical lever (Figure 1) operates by reflecting a laser beam off the cantilever. Angular deflection of the cantilever causes a twofold larger angular deflection of the laser beam. The reflected laser beam strikes a position-sensitive photodetector consisting of two side-by-side photodiodes. The difference between the two photodiode signals indicates the position of the laser spot on the detector and thus the angular deflection of the cantilever. Image acquisition times is of about one minute.
Atomic force microscope, University of Konstanz (May 1991) The ability of AFM to image at atomic resolution, combined with its ability to image a wide variety of samples under a wide variety of conditions, has created a great deal of interest in applying it to the study of biological structures. Images have appeared in the literature showing DNA, single proteins, structures such as gap junctions, and living cells.
• Linear tomography (~1930-th) Moving X-ray point source O1
O2
O3
Focal plane
Image 3
Image 2
Image 1
Moving stage with a X-ray sensor Schematic diagram of linear tomography. Due to the synchronous movement of the X-ray source and X-ray sensor, certain plane cross-section of the object is always projected in the same place of the sensor while others are projected with a displacement and therefore will appear blurred in the resulting image.
Application in dentistry
Laminography
The principle of laminography (http://lca.kaist.ac.kr/Research/2000/X_lamino.html) X-ray point source moving in the source plane over a circular trajectory projects object onto X-ray detector plane. The detector moves synchronously to the source in such a way as to secure that a specific object layer is projected on the same place on the detector array for whatever position of the source. The plane of this selected layer is called “focal plane’. Projections of other object layers located above or beneath of the “focal plane” will, for different position of the source, be displaced. Therefore if one sums up all projections obtained for different positions of the source, projections of the focal plane layers will be accumulated coherently producing a sharp image of this layer while other layers projected with different displacement in different projections will produce a blurred background image. The more projections are available the lower will be the contribution of this background into high frequency components of the output image.
Projection
Accumulated image
Projection
Accumulated image
Illustration of restoration of different layers of a printed circuit board
TRANSFORM IMAGING TECHNIQUES All above imaging devices belong to a class of direct image plane imaging devices. They produce images that can be directly viewed by eyes. Fundamental drawbacks of direct image plane imaging techniques is that they require direct access to individual locations of objects to be resolved Transform imaging devices collect, in a certain form, information needed for image “reconstruction” rather then directly object images. In transform imaging, image information retrieval and image formation (reconstruction) for display are essentially separated. Probably, the very first example of indirect imaging method was that of X—ray crystallography (Max Von Laue, 1912, Nobel Prize 1914-1918) In 1912 Max von Laue and his two students (Walter Friedrich and Paul Knipping) demonstrated the wave nature of X-rays and periodic structure of crystals by observing the diffraction of X-rays from crystals of zinc sulfide. This discovery ended up in crystallography as a new imaging technique. In crystallography, geometrical parameters and intensity distribution of the pattern of diffraction spots is used for calculation of spatial distribution and density of atoms in crystals.
Discovery of diffraction of X-rays had a decisive value in the development of physics and biology of XX-th century. One of the most remarkable scientific achievements that is based on X-ray crystallography was discovery by J. Watson and F. Crick of spiral structure of DNA (Nobel Prize, 1953)
• Holography (1948, D. Gabor, The Nobel Prize, 1971)
Invention of holography by D. Gabor (1948) was motivated by the desire to improve resolution power of electron microscopes that was limited by the fundamental limitations of the electron optics. The term “holography” originates from Greece word “holos” (ηωλωσ). By this, inventor of holography intended to emphasize that in holography full information regarding light wave, both amplitude and phase, is recorded by means of interference of two beams, object and reference one. Due to the fact that at that time sources of coherent electron radiation were not available, Gabor carried out model optical experiments to demonstrate the feasibility of the method. However, powerful sources of coherent light were also not available at the time, and holography remained an “optical paradox” until the invention of lasers. The very first implementation of holography were demonstrated in 1961 by radio-engineers E. Leith a nd J. Upatnieks in Michigan University and by Physicist Yu. Denisyuk in State Optical Institute, Sanct Petersburg, Russia. In holography, interference pattern between optical waves reflected by or transmitted through object and a special “reference” wave is recorded. Recording medium
Source of coherent light Object beam
Principle of recording holograms
Object
For reconstruction of hologram, hologram is illuminated by the same “reference” beam used for recording.
Reference beam
Mirror
Principle of hologram playback
Source of coherent light
Recording medium
Virtual object (real)
Reconstructing beam
Mirror Virtual object (imaginary)
• Optical information processing (Marechal, VanderLugt, 1964-66) Invention of lasers and holography stimulated works on optical information processing based on capability of by capability of optical lenses perform image Fourier Transforms with a speed of light.
Input image
Fourier lens
F
Fourier lens
Spatial filter
F
F
Output image
F
Coherent illumination
Optical system for image restoration
Input image Fourier lens
F
F
Matched filter: Target object h l
Fourier lens
F
Coherent illumination
Vander Lugt optical correlator
Correlation plane
F
• Digital holography (1968-1971) In the end of 1960-th beginning 1970-th it was suggested (A. Lohmann, J. Goodman, T. Huang) to use digital computers for reconstruction and synthesis of holograms and replacement optical information processing.
− Computer reconstruction of holograms − Computer synthesis of holograms for 3-D visualization − Computer synthesis of diffractive optical elements for optical image processing and optical metrology Digital holography reflects, in the most purified way, the informational pith and marrow of holography and imaging which motivated two the most famous inventors in holography, D. Gabor and Yu.N. Denisyuk
Beam spatial filter
Reference beam
Collimator Lens
Hologram sensor: Digital photographic camera
Microscope
Laser Object table
Digital reconstruction of electronically recorded holograms
Object beam
Computer
Latest development: digital holographic microscopy (end of 1990-th)
•
Synthetic aperture radar (C.Wiley, USA, 1951)
In synthetic aperture radar imaging, amplitude and phase of radio waves reflected by the object is recorded in course of plain flight around the object. These flight data are then used for reconstruction of wave reflectivity distribution over the object surface. The reconstruction is carried out either optically, or, presently, in digital computers. It is not accidentally that E. Leith and Yu. Upatnieks who produced first optical laser hologram were radio engineers that have been working on synthetic aperture radars, a highly classified subject at that time.
One of most recent example of application of SAR imaging: Radar map of Venus
If the thick clouds covering Venus were removed, how would the surface appear? Using an imaging radar technique, the Magellan spacecraft was able to lift the veil from the Face of Venus and produce this spectacular high resolution image of the planet's surface. Red, in this false-color map, represents mountains, while blue represents valleys. This 3-kilometer resolution map is a composite of Magellan images compiled between 1990 and 1994. Gaps were filled in by the Earth-based Arecibo Radio Telescope. The large yellow/red area in the north is Ishtar Terra featuring Maxwell Montes, the largest mountain on Venus. The large highland regions are analogous to continents on Earth. Scientists are particularly interested in exploring the geology of Venus because of its similarity to Earth.
•
“Coded” aperture (multiplexing) techniques (1970-th)
Pinhole camera (camera obscura) has a substantial advantage over lenses - it has infinite depth of field, and it doesn't suffer from chromatic aberration. Because it doesn't rely on refraction, pinhole camera can be used to form images from X-ray and other high energy sources, which are normally difficult or impossible to focus. Source of irradiation Pinhole camera
Image plane detector
The biggest problem with pinhole cameras is that they let very little light through to the film or other detector. This problem can be overcome to some degree by making the hole larger, which unfortunately leads to a decrease in resolution. The smallest feature which can be resolved by a pinhole is approximately the same size as the pinhole itself. The larger the hole, the more blurred the image becomes. Using multiple, small pinholes might seem to offer a way around this problem, but this gives rise to a confusing montage of overlapping images. Nonetheless, if the pattern of holes is carefully chosen, it is possible to reconstruct the original image with a resolution equal to that of a single hole.
Coding mask
m( x , y )
Detector array
b( x , y )
In coded aperture imaging, image projections obtained through a set of special binary masks are recorded and used for image reconstruction carried out in computers
•
Computer tomography (Hounsfield,
1973, The Nobel Prize,
~1980) Parallel beam of X-rays
y
Obj ( x , y )
x
ϑ X-ray sensitive line sensor array
ξ Pr oj (ϑ ,ξ )
Schematic diagram of parallel beam projection tomography In computer tomography, a set of object’s projections taken a different observation angles is measured and used for subsequent reconstruction of the object. Computer tomography had become the first full scale example of digital imaging
Surface rendering of a fly head reconstructed using a SkyScan micro-CT scanner Model L1072 (Advanced imaging, July 2001, p. 22)
• MRI : Magnetic resonance tomography (Felix Bloch and Edward Purcell, The Nobel Prize, 1952, for the discovery of the magnetic resonance phenomenon in 1946; Richard Ernst, The Nobel Prize in chemistry, 1991 for his achievements in pulsed Fourier Transform NMR and MRI; Paul C. Lautenbur and Sir Peter Mansfield, UK, the Nobel Prize in physiology and medicine, 2003) . Magnet and “gradient” coils
Strong magnetic field
z
RF inductor
Object y
x
Reconstruction and display
RF receiver
RF impulse generator
Schematic diagram of NMR imaging MRI is based on the principles of nuclear magnetic resonance (NMR), a spectroscopic technique capable of obtaining microscopic chemical and physical information about molecules. An effect is observed when an atomic nucleus is exposed to radio waves in the presence of a magnetic field. A strong magnetic field causes the magnetic moment of the nucleus to precess around the direction of the field, only certain orientations being allowed by quantum theory. A transition from one orientation to another involves the absorption or emission of a photon, the frequency of which is equal to the precessional frequency. With magnetic field strengths customarily used, the radiation is in the radio-frequency band. If radio-frequency radiation is supplied to the sample from one coil and is detected by another coil, while the magnetic field strength is slowly changed, radiation is absorbed at certain field values, which correspond to the frequency difference between orientations. An NMR spectrum consists of a graph of field strength against detector response. This provides information about the structure of molecules and the positions of electrons within them, as the orbital electrons shield the nucleus and cause them to resonate at different field strengths. ( adopted from The Macmillan Encyclopedia 2001, © Market House Books Ltd 2000)
Nobel prizes for new imaging devices and principles of imaging •
Wilhelm Conrad Röntgen, Germany, Munich University , Munich, Germany b.1845,
•
Gabriel Lippmann, France, Sorbonne University, Paris, France b.1845 (in Hollerich, Luxembourg), d.1921: The Nobel Prize in Physics 1908 "for his method of reproducing colours photographically based on the phenomenon of interference"
•
Max von Laue, Germany, Frankfurt-on-the Main University , Frankfurt-on-the Main,
•
Patrick Maynard Stuart Blackett, United Kingdom, Victoria University,
•
Cecil Frank Powell, United Kingdom, Bristol University, Bristol, United Kingdom
•
Frits (Frederik) Zernike, the Netherlands, Groningen University , Groningen, the
•
Donald Arthur Glaser, USA, University of California , Berkeley, CA, USA b.1926.
•
Dennis Gabor, United Kingdom, Imperial College of Science and Technology London,
•
Allan M. Cormack, USA, Tufts University Medford, MA, USA, b.1924 (in
•
Godfrey N. Hounsfield United Kingdom Central Research Laboratories, EMI, London, United Kingdom b.1919. The Nobel Prize in Physiology or Medicine, 1979 "for the development of computer assisted tomography " Ernst Ruska, Germany Fritz-Haber-Institut der Max-Planck- Gesellschaft, Berlin, b.1906, d.1988. The Nobel Prize in Physics 1986 "for his fundamental work in electron optics, and for the design of the first electron microscope"
•
•
d.1923. The Nobel Prize in Physics 1901 "in recognition of the extraordinary services he has rendered by the discovery of the remarkable rays subsequently named after him"
Germany b.1879, d.1960:The Nobel Prize in Physics 1914 "for his discovery of the diffraction of X-rays by crystals" Manchester, United Kingdom b.1897, d.1974The Nobel Prize in Physics 1948 "for his development of the Wilson cloud chamber method, and his discoveries therewith in the fields of nuclear physics and cosmic radiation" b.1903 d.1969. The Nobel Prize in Physics 1950 "for his development of the photographic method of studying nuclear processes and his discoveries regarding mesons made with this method" Netherlands, b.1888, d.1966. The Nobel Prize in Physics 1953 "for his demonstration of the phase contrast method, especially for his invention of the phase contrast microscope" The Nobel Prize in Physics 1960 "for the invention of the bubble chamber” United Kingdom b.1900 (in Budapest, Hungary), d.1979, The Nobel Prize in Physics 1971 "for his invention and development of the holographic method" Johannesburg, South Africa) d.1998
Gerd Binnig, Germany, b.1947, , IBM Zurich Research Laboratory, Switzerland Heinrich Rohrer , Switzerland, b.1933, IBM Zurich Research Laboratory, Switzerland. The Nobel Prize in Physics 1986 "for their design of the scanning tunneling microscope“
•
Paul C. Lautenbur, Peter Mansfield, UK. The Nobel Prize 2003 in Physiology and Medicine ”for their discoveries concerning magnetic resonance imaging”.
Digital imaging and image processing: the highest level of the evolution of imaging techniques New qualities that are brought to imaging systems by digital computers and processors: -
Flexibility and adaptability. The most substantial advantage of digital computers as compared with analog electronic and optical information processing devices is that no hardware modifications are necessary to reprogram digital computers to solving different tasks. With the same hardware, one can build an arbitrary problem solver by simply selecting or designing an appropriate code for the computer. This feature makes digital computers also an ideal vehicle for processing image signals adaptively since, with the help of computers, they can adapt rapidly and easily to varying signals, tasks and end user requirements.
-
Digital computers integrated into imaging systems enable them to perform not only element-wise and integral signal transformations such as spatial and temporal Fourier analysis, signal convolution and correlation that are characteristic for analog optics but any operations needed. This removes the major limitation of optical information processing and makes optical information processing integrated with digital signal processing almost almighty.
-
Acquiring and processing quantitative data contained in images as signals, and connecting imaging systems to other informational systems and networks is most natural when data are handled in digital form. In the same way as in economics currency is a general equivalent, digital signals are general equivalent in information handling. A digital signal within the computer that represents, so to say, purified information carried by image signals deprived of its physical integument. Thanks to its universal nature, the digital signal is an ideal means for integrating different informational systems.
The only limitations of digital imaging and image processing are memory and processing speed capacities of computers.
IMAGE PROCESSING: SOLVING a GIGO PROBLEM
USER
OBJECTS of STUDY
IMAGING DEVICE
Raw
IMAGE
IMAGE (Garbage In)
PROCESSOR
(Gold Out)
Three types of end users in image processing: • • •
“Collective” user (as in commercial photography, TV broadcasting, Multimedia) Expert user (as in air and space photo reconnaissance, radiology diagnostics, etc.) Automatic devices (computer vision)
If it’s green or it wriggles, it’s biology If it doesn’t work, it’s physics To err is human, but to really fool things requires a computer Murphy’s handy guide to modern science
IMAGE PROCESSING TASKS User Image processing:
Image formation (image reconstruction) Image perfecting (“Image restoration”) Interactive image interpretation (“Image preparation”, ”Image enhancement”) Image quantification (parameter estimation) Automated image analysis Image modeling (“Virtual imaging”) Image coding for transmission and storage Simulating imaging devices
Collective Individual Automata user user expert • • • •
•
•
•
•
• •
•
•
•
•
•
•
•
Digital Image Processing: Applications An ounce of application is worth of a ton of abstraction Booker’s law, in: A. Bloch, Murphy’s Law, Price Stern Sloan,L.A.,1977 Es gibt nichts Praktischeres als eine gute Theorie (There is nothing more practical then a good theory) Saying Grau, teurer Freund, ist alle Theorie And gruen des Lebens goldner Baum (Gray, dear friend, is every theory And green the golden tree of life) Goethe, Faust La precision n’est pas la fidelite (Accuracy is not fidelity) H. Matisse
Syllabus of the course: • Image digitization and coding • Image interpolation, resampling and geometrical transforms. • Statistical image and noise models and texture synthesis and analysis. • Image quantification and parameter estimation techniques • Object detection and localization in images. Accuracy and reliability of the localization. Adaptive and local adaptive filters for reliable target localization. • Image reconstruction, perfecting and restoration. Optimal, adaptive and local adaptive linear and rank filters. • Interactive image processing and enhancement. • Multi component and multi modal image processing. • Efficient computational algorithms and parallel neuro-morphic networks Text books: 1. L. Yaroslavsky, Digital Holography and Digital Image processing: Principles, Methods, Algorithms, Kluwer Scientific Publishers, 2004 2. L. Yaroslavsky, M. Eden. Fundamentals of Digital Optics. Birkhauser, Boston, 1996 3. L. Yaroslavsky. Digital Picture Processing. An Introduction, Springer Verlag, Heidelberg, New York, 1985
Additional references 1. 2. 3. 4. 5. 6. 7.
L. Yaroslavsky, Digital Signal Processing in Optics Holography, Moscow, Radio I Svyaz’), 1987 (In Russian). L. P. Yaroslavsky, N. S. Merzlyakov, Digital Holography, Nauka Publsh., Moscow, 1982 (In Russian). L. P. Yaroslavskii, N. S. Merzlyakov, Methods of Digital Holography, Consultant Bureau, N.Y., 1980 L. Yaroslavsky, Introduction to Digital Image processing, Moscow, Sov. Radio, Moscow, 1979 (In Russian). J. S. Lim, Two-dimensional Signal and Image Processing, Prentice Hall, New Jersey, 1990. R. C. Gonzalez, R. E. Woods, Digital Image Processing, 2-nd ed. Prentice Hall, Inc., 2002 R. N. Bracewell, Two-Dimensional Imaging, Prentice Hall Int. Inc., 1995
