Towards Automated Creation of Image Interpretation Systems

2003 ◽  
pp. 653-665 ◽  
Author(s):  
Ilya Levner ◽  
Vadim Bulitko ◽  
Lihong Li ◽  
Greg Lee ◽  
Russell Greiner
Keyword(s):  
Image Interpretation ◽  
Interpretation Systems

System architectures for interactive knowledge-based image-interpretation

1988 ◽  
Vol 324 (1579) ◽  
pp. 457-465 ◽  
Keyword(s):  
Software Architecture ◽  
Complete System ◽  
Image Interpretation ◽  
System Architectures ◽  
Low Level ◽  
Knowledge Based ◽  
Systematic Strategy ◽  
High Level ◽  
Interpretation Systems

We discuss hardware and software architecture for automated image-interpretation. The importance of considering the complete system is emphasized leading in particular to the conclusion that high-level and low-level processing are intimately linked. We present arguments to support the idea that automated image-interpretation systems should be knowledge-based and interactive. We attempt to identify the main architectural problems which such systems must address and outline a systematic strategy for acquiring, structuring and using knowledge.

Observation of biological macromolecules using various electron microscopic methods

1978 ◽  
Vol 36 (2) ◽  
pp. 170-171
Author(s):  
Mitsuo Ohtsuki ◽  
Michael Sogard
Keyword(s):  
Electron Microscope ◽  
Phase Contrast ◽  
Image Interpretation ◽  
Density Lipoprotein ◽  
Biological Macromolecules ◽  
Contrast Effects ◽  
Biological Structure ◽  
Electron Microscopic ◽  
Structural Investigations ◽  
Staining Techniques

Structural investigations of biological macromolecules commonly employ CTEM with negative staining techniques. Difficulties in valid image interpretation arise, however, due to problems such as variability in thickness and degree of penetration of the staining agent, noise from the supporting film, and artifacts from defocus phase contrast effects. In order to determine the effects of these variables on biological structure, as seen by the electron microscope, negative stained macromolecules of high density lipoprotein-3 (HDL3) from human serum were analyzed with both CTEM and STEM, and results were then compared with CTEM micrographs of freeze-etched HDL3. In addition, we altered the structure of this molecule by digesting away its phospholipid component with phospholipase A2 and look for consistent changes in structure.

Simulated Dark-Field Electron Micrographs of Point Defects and Organometallics

1975 ◽  
Vol 33 ◽  
pp. 200-201
Author(s):  
William Krakow
Keyword(s):  
Electron Micrographs ◽  
Point Defects ◽  
Structure Determination ◽  
Atomic Structure ◽  
Image Interpretation ◽  
Dark Field ◽  
Field Electron ◽  
Dark Field Microscopy ◽  
Dark Field Electron

Tilted beam dark-field microscopy has been applied to atomic structure determination in perfect crystals, several synthesized molecules with heavy atcm markers and in the study of displaced atoms in crystals. Interpretation of this information in terms of atom positions and atom correlations is not straightforward. Therefore, calculated dark-field images can be an invaluable aid in image interpretation.

A method for direct measurement of specimen noise in HREM images

1993 ◽  
Vol 51 ◽  
pp. 458-459
Author(s):  
Sidnei Paciornik ◽  
Roar Kilaas ◽  
Ulrich Dahmen ◽  
Michael Adrian O'Keefe
Keyword(s):  
Random Noise ◽  
Image Interpretation ◽  
Thin Foil ◽  
High Resolution Electron Microscopy ◽  
Black Dot ◽  
Atomic Column ◽  
Amorphous Oxide ◽  
Resolution Electron ◽  
Imaging Conditions ◽  
Image Simulations

High resolution electron microscopy (HREM) is a primary tool for studying the atomic structure of defects in crystals. However, the quantitative analysis of defect structures is often seriously limited by specimen noise due to contamination or oxide layers on the surfaces of a thin foil.For simple monatomic structures such as fcc or bcc metals observed in directions where the crystal projects into well-separated atomic columns, HREM image interpretation is relatively simple: under weak phase object, Scherzer imaging conditions, each atomic column is imaged as a black dot. Variations in intensity and position of individual image dots can be due to variations in composition or location of atomic columns. Unfortunately, both types of variation may also arise from random noise superimposed on the periodic image due to an amorphous oxide or contamination film on the surfaces of the thin foil. For example, image simulations have shown that a layer of amorphous oxide (random noise) on the surfaces of a thin foil of perfect crystalline Si can lead to significant shifts in image intensities and centroid positions for individual atomic columns.

Molecular Views of Ice-Embedded Lumbricus Terrestris Erythrocruorin Obtained By Invariant Classification

1990 ◽  
Vol 48 (1) ◽  
pp. 450-451
Author(s):  
Michael schatz ◽  
Joachim Jäger ◽  
Marin van Heel
Keyword(s):  
Image Interpretation ◽  
Lumbricus Terrestris ◽  
Reference Image ◽  
Large Set ◽  
Multivariate Statistical ◽  
Data Set ◽  
Low Contrast ◽  
Negative Stain ◽  
Reference Images ◽  
Statistical Resolution

Lumbricus terrestris erythrocruorin is a giant oxygen-transporting macromolecule in the blood of the common earth worm (worm "hemoglobin"). In our current study, we use specimens (kindly provided by Drs W.E. Royer and W.A. Hendrickson) embedded in vitreous ice (1) to avoid artefacts encountered with the negative stain preparation technigue used in previous studies (2-4).Although the molecular structure is well preserved in vitreous ice, the low contrast and high noise level in the micrographs represent a serious problem in image interpretation. Moreover, the molecules can exhibit many different orientations relative to the object plane of the microscope in this type of preparation. Existing techniques of analysis requiring alignment of the molecular views relative to one or more reference images often thus yield unsatisfactory results.We use a new method in which first rotation-, translation- and mirror invariant functions (5) are derived from the large set of input images, which functions are subsequently classified automatically using multivariate statistical techniques (6). The different molecular views in the data set can therewith be found unbiasedly (5). Within each class, all images are aligned relative to that member of the class which contributes least to the classes′ internal variance (6). This reference image is thus the most typical member of the class. Finally the aligned images from each class are averaged resulting in molecular views with enhanced statistical resolution.

Sign in / Sign up

Export Citation Format

Share Document