Biomedical subjects

E Peli

Publications and source records attributed to E Peli.

At least 19 recordsLinked to original sources

PubMed1992

Limitations of image enhancement for the visually impaired.

Image enhancement as an aid for the visually impaired may improve visibility of TV programs and provide portable visual aid. This paper describes the current techniques for image enhancement and their underlying models. The limitations of the various techniques and of potential methods of implementation are high-lighted. Initial work in this area was based on a linear model. The finite dynamic range available in the video display and contamination of the enhanced image by high spatial frequency noise limited the model's usefulness. I propose a method to address some limitations of the original model that considers the nonlinear response of the visual system and requires enhancement of subthreshold spatial information only. This modification may increase the dynamic range available by decreasing the range previously used by the linear models to enhance visible details. However, for the modified technique to be most effective, the enhancement has to be continuously tuned, based on the patient's visual loss and the spatial frequency content of the displayed images. The implications of these limitations for the potential implementation in TV are discussed. Implementation of an image-enhancing visual aid in a head-mounted, binocular, full-field, virtual vision device may cause substantial difficulties. Patient adaptation may be difficult due to head movement and interaction of the vestibular system response with the head-mounted display. An alternate, bioptic design is proposed in which the display is positioned above or below the line of sight to be examined intermittently, possibly in a freeze-frame mode. Such implementation is also likely to be less expensive, enabling more users access to the device.

Audiovisual Aids↗

PubMed1992

Differences in tests of aniseikonia.

The New Aniseikonia Test (NAT), a hand-held direct-comparison test using red/green anaglyphs, has several potential advantages as a screener. We compared the validity of the NAT to that of the Space Eikonometer in three experiments: (1) aniseikonia was induced by calibrated size lenses in a double-blind study of 15 normal subjects; (2) habitual aniseikonia was measured with both instruments in four patients; and (3) eight of the normal subjects were retested with a computer-video simulation of the NAT. The NAT underestimated induced aniseikonia by a factor of 3 in the normal subjects and underestimated habitual aniseikonia in four patients. The Space Eikonometer correctly measured the magnitude of induced aniseikonia in the normal subjects. The simulation test did not show underestimation in the eight normal subjects. We could not attribute the NAT's underestimation of aniseikonia to the red/green anaglyph method, printing error, psychophysical method, or the direct-comparison test format. We speculate that the NAT induces a different sensory fusion response to aniseikonia than do the other tests, and that this altered sensory fusion response diminishes measured aniseikonia. We conclude that the NAT is not a valid measure of aniseikonia.

Adult↗

PubMed1991

Image analysis of changes in drusen area.

Computerized image processing was used to analyze color fundus photographs of 11 patients (22 eyes) with macular drusen who were followed for more than 2 years (mean follow-up, 4.7 years). Significant changes over time (more than +/- 20% of baseline area) were measured in the surface area of macular drusen in 18 of 22 (82%) eyes. An increase in the drusen area was associated significantly with eyes with mostly hard drusen and an initially smaller drusen area; but a decrease was associated with eyes with mostly soft drusen and an initially larger drusen area (P less than 0.01). The mean absolute rate of change in the drusen area was more than twice as great in eyes with mostly soft drusen compared with those with mostly hard drusen (P less than 0.05). All eye pairs studied showed a concomitant increase or decrease in the drusen area.

Aged↗

PubMed1991

Multiresolution, error-convergence halftone algorithm.

A new halftone algorithm is described. The algorithm is designed for implementation on a parallel architecture in order to provide fast, progressive coding of moderate-resolution images. The design is based on a multiresolution, hierarchical, pyramidal structure. At each pyramid level, the binarized image is compared with the original, gray-tone image over a successively larger window of pixels for calculation of a weighted averaged error. Within each level, selected binarized pixels are tested for possible changes in the binary assignment. The binary assignment is changed if the change results in a lower average error over the entire window. Varying the selection of test pixels can cause the same process to provide clustered-dot patterns and dithering. A comparison of performance with the best implementation of the error-propagation algorithm is presented visually. Quality is compared also in terms of isotropy of the texture and the appropriate blue-noise characteristics in areas of uniform gray tone. The benefits of this algorithm are realized with moderate-resolution display of the order of 512 dots X 512 dots. The processing can be carried out on smaller blocks since the results can be combined without any visible seams or edge effects.

Algorithms↗

PubMed1991

Effect of luminance on suprathreshold contrast perception.

Perceived contrast was measured under natural viewing conditions with the use of contrast-matching and magnitude-estimation paradigms and found to be independent of luminance over a range of luminances from 37.5 down to 8 cd/m2. However, this contrast constancy broke down when the dimmer target was below 8 cd/m2. The perceived contrast of the dimmer target then fell below that expected from contrast constancy. The extended range of contrast constancy previously reported [J. Physiol. 252, 627 (1975); Vision Res. 16, 1419 (1976)] has been thought to imply neural mechanisms with unlimited constancy, but these researchers permitted differential adaptation to the brighter and dimmer targets, which were seen haploscopically (by different eyes). As our natural-viewing procedure ensured that both bright and dim targets were presented to retinal areas in a roughly constant state of adaptation, our failure to find extended contrast constancy implies an important limitation on the neural processing of contrast.

Adaptation, Ocular↗

PubMed1991

Image invariance with changes in size: the role of peripheral contrast thresholds.

The appearance of objects generally does not change with changes in the size of their retinal image that occur as the distance from the observer increases or decreases. Contrast constancy ensures this invariance for suprathreshold image features, but fully robust size invariance also requires invariance at threshold, so that near-threshold image features do not appear or disappear with distance changes. Since the angular size and the eccentricity of image features covary with distance changes, the threshold requirement for invariance could be satisfied approximately if contrast thresholds were to vary as the product of the spatial frequency and the eccentricity from the fovea. This model fits contrast thresholds for orientation identification over spatial frequencies of 1-16 cycles/deg and for retinal eccentricities of as much as 23 deg. Contrast detection thresholds from six different studies conform to this model over an even wider range of spatial frequencies and retinal eccentricities. The fitting variable, the fundamental eccentricity constant, was similar for all three studies that measured detection along the horizontal meridian and was higher for the orientation identification contrast thresholds along the same meridian. The eccentricity constant from studies that measured detection along the vertical meridian was higher than the constant calculated for the horizontal meridian and lower than the eccentricity constant for chromatic isoluminance gratings. Our model and these results provide new tools for analyzing the visibility of displays and for designing equal-visibility or variable-visibility displays.

Contrast Sensitivity↗

PubMed1991

Image enhancement for the visually impaired. Simulations and experimental results.

Digital image enhancement has been proposed as an aid for the visually impaired. The capability of two enhancement techniques to improve recognition of images by patients with central scotoma or cataracts was evaluated using image-processing simulations and direct patient testing. Enhancements and simulations were based on measurements of contrast sensitivity loss for patients with macular disease. Contrast sensitivity loss was measured using Gabor-type localized stimuli and paradigms that are appropriate for analyzing form perception. The simulations using the contrast sensitivity data suggested that patients with moderate visual loss (20/70-20/200) may have difficulty recognizing faces and may benefit from enhancement by both of the techniques used. Ability to recognize celebrities from enhanced images improved for 39 of the 46 patients tested. The improvement was significant (P less than 0.05) for 16 of the 38 patients with central visual loss and for 3 of 8 patients with anterior segment media opacities tested. The simulations suggest that the benefits of image enhancement may be similar or even greater for recognition of other types of images.

Adult↗

PubMed1990

Contrast in complex images.

The physical contrast of simple images such as sinusoidal gratings or a single patch of light on a uniform background is well defined and agrees with the perceived contrast, but this is not so for complex images. Most definitions assign a single contrast value to the whole image, but perceived contrast may vary greatly across the image. Human contrast sensitivity is a function of spatial frequency; therefore the spatial frequency content of an image should be considered in the definition of contrast. In this paper a definition of local band-limited contrast in images is proposed that assigns a contrast value to every point in the image as a function of the spatial frequency band. For each frequency band, the contrast is defined as the ratio of the bandpass-filtered image at the frequency to the low-pass image filtered to an octave below the same frequency (local luminance mean). This definition raises important implications regarding the perception of contrast in complex images and is helpful in understanding the effects of image-processing algorithms on the perceived contrast. A pyramidal image-contrast structure based on this definition is useful in simulating nonlinear, threshold characteristics of spatial vision in both normal observers and the visually impaired.

Contrast Sensitivity↗

PubMed1988

Use of circularly polarized light in fundus and optic disc photography.

Circularly polarized light was used in fundus and optic disc photography to reduce reflected glare and photographic artifacts originating from the ocular media and the lenses of the fundus camera. A standard fundus camera was used, with circular polarizers placed in the illumination and observation paths. The quality of fundus photography was improved significantly in eyes with intraocular lenses, cataract, vitreous haze, or high myopia as well as in normal eyes, in fundus photography of the retinal periphery, and in optic disc photography.

Fundus Oculi↗

PubMed1987

Enhancement of fundus photographs taken through cataracts.

Image-enhancement techniques can aid the clinician in evaluating fundus pictures taken through cataracts. The degradation of the fundus image by cataracts has been described as low-pass filtering. Means to partially overcome this degradation using homomorphic filtering and adaptive enhancement are presented. The clinical value of these enhancement techniques is demonstrated with two cases of progressive glaucoma and cataracts.

Cataract↗

PubMed1987

Nerve fiber layer photography. A comparative study.

Observation of the retinal nerve fiber layer with red-free light is an important means of evaluating optic nerve integrity. Many investigators have attempted to increase the usefulness of the technique by improving photographic procedures. Most reports comparing various parameters neither indicate the method of comparison nor give quantitative measure of the improvement gained with different films, filters, cameras, or processing procedures. We have compared in a masked fashion the parameters most recently recommended in the literature. Twelve trained observers compared 12 masked pairs of nerve fiber layer photos taken from the same eye in the same manner on two cameras and preferred those taken with a Canon CF-60Z camera 69% of the time and those taken with a Zeiss FF-111 17% of the time (this difference was statistically significant, P less than 0.01). Fourteen observers compared 14 pairs of photographs and judged Panatomic-X film to be superior to Plus-X film 51% of the time and both films to be of the same quality 25% of the time (this difference was also significant P less than 0.05). Fifteen observers compared 12 pairs and preferred the Standard SE-40 blue filter 2:1 over the green Spectrotech 540 filter. This difference was not statistically significant, and the preference was found to be dependent on fundus pigmentation and clarity of the media. The blue filter gave better contrast for the nerve fiber layer in patients with lightly pigmented fundi, and use of the green filter resulted in less light scatter in patients with ocular media opacities.

Evaluation Studies as Topic↗

PubMed1987

Reading with a macular scotoma. II. Retinal locus for scanning text.

To elucidate how patients with macular scotomas use residual functional retina for inspecting visual detail and reading, we tested three patients with dense macular scotomas using a scanning laser ophthalmoscope that allows an examiner to view and record stimuli on the retina while the patient views them. Using standard psychophysical techniques, we determined the retinal position of scotomas, the retinal areas used for fixating and inspecting acuity targets, and the retinal area used for reading simple, three-letter, nonsense syllables. We found that each patient used a single, idiosyncratic retinal area, immediately adjacent to the scotoma, for fixating, inspecting acuity targets, and scanning simple, nonsense-syllable text. This preferred retinal locus (PRL) was at different retinal eccentricities (relative to the foveola) for each patient and was not always as close as possible to the foveola. There appears to be no simple rule by which patients "select" a particular PRL. Plots of text placement on the retina revealed considerable differences in patients' abilities to execute an orderly text scan. Two patients read text more rapidly with a novel retinal area than with their PRL, suggesting that the PRL may not be optimal for text reading.

Eye Movements↗

PubMed1986

Drusen measurement from fundus photographs using computer image analysis.

Drusen are yellowish deposits at the level of the retinal pigment epithelium and are frequently associated with age-related maculopathy (ARM). Drusen often change in size and number over time and may be followed by atrophic or exudative macular degeneration. A quantitative method to measure the development of drusen is needed for controlled studies of the natural history, prognosis, and treatment of ARM. An objective method is described using computer image analysis of fundus photographs for the detection and measurement of drusen. This technique enables us to measure both the area of drusen in the macula and the changes in the drusen pattern over time. Evaluation of repeated photographs showed reproducibility of 6.1%, whereas the reproducibility of processing photographic duplicates was 2.3%. Digitization with a high-quality linear array solid state camera did not change reproducibility significantly.

Fundus Oculi↗

PubMed1986

Control of eye movement with peripheral vision: implications for training of eccentric viewing.

A review of peripheral visual control of eye movements suggests that such control may be easier if the direction of the extrafoveal locus used is orthogonal rather than radial to the direction of target motion. When the extrafoveal locus lies parallel (radial) to the direction of target or eye motion, artificial feedback is usually required to avoid reflexive foveation. These findings have important implications for training eccentric viewing in low vision patients with central scotomata.

Aged↗

PubMed1986

Blink vergence in an antimetropic patient.

A patient with uncorrected antimetropia was found to attain motor fusion through blinking. Although this patient was also able to attain motor fusion through saccadic vergence and slow fusional vergence, he usually relied on blink vergence. In this patient, blink vergence was an efficient alternative to slow fusional vergence.

Adult↗

PubMed1986

Computerized enhancement of retinal nerve fiber layer.

We investigated a number of image processing techniques for enhancing the retinal nerve fiber layer (NFL) in fundus photographs taken with red-free light. Wedge-type defects within the NFL were enhanced best by contrast enhancement algorithms such as histogram modification and extremum sharpening. Narrow, slitlike NFL defects can be enhanced by directional enhancement techniques. Normal NFL features were enhanced best by high-pass filtering techniques such as homomorphic filtering and adaptive enhancement. The enhanced normal NFL may be useful in the analysis of diffuse NFL loss. All of these successful enhancements of normal and defective features indicate that enhancing fundus photographs with digital image processing techniques should facilitate early detection of optic nerve damage from glaucoma and other diseases.

Computers↗

PubMed1986

Ophthalmic applications of circular polarizers.

Circular polarizing filters have recently been introduced in several ophthalmic applications. These applications include clinical instruments for ophthalmoscopy, slit-lamp specular microscopy, and fundus photography, as well as consumer products such as 3-D movie projection and viewing systems and glare-control screens for computer visual display terminals. The nature of circularly polarized light and mode of operation of the circular polarizing filters in various applications are discussed. In view of the advantages of circular polarizers in vectographic applications, the current linearly polarized vectographic systems used for binocular testing and visual training should be replaced with circular polarizing systems.

Computers↗

PubMed1986

Reading with a macular scotoma. I. Retinal location of scotoma and fixation area.

To investigate how patients with macular scotomas use residual functional retinal areas to inspect visual detail, a scanning laser ophthalmoscope (SLO) was used to map the retinal locations of scotomas and areas used to fixate. Three patients with dense macular scotomas of at least 20 months duration and with no explicit low vision training were tested. SLO stimuli were produced by computer modulation of the scanned laser beam, and could be placed on known retinal loci by direct observation of the retina on a television monitor. Videotaped SLO images were analyzed to produce retinal maps that are corrected for shifts of stimulus position due to fixational eye movement, thus showing the true retinal locations of scotomas and fixation loci. Major findings were as follows: 1) each patient used a single, idiosyncratic retinal area, immediately adjacent to the scotoma to fixate, and did not attempt to use the nonfunctional foveola, 2) fixation stability with the eccentric fixation locus was as good as, or better than, that of ocularly normal subjects trying to fixate at comparable eccentricities, 3) fixation stability was not systematically related to clinical visual acuity, and 4) there is good agreement as to the shape and overall size of SLO and standard clinical tangent screen scotoma maps for these three patients.

Adult↗