The detection of deviation from straightness in lines

Image quality assessment based on inter-patch and intra-patch similarity

In this paper, we propose a full-reference (FR) image quality assessment (IQA) scheme, which evaluates image fidelity from two aspects: the inter-patch similarity and the intra-patch similarity. The scheme is performed in a patch-wise fashion so that a quality map can be obtained. On one hand, we investigate the disparity between one image patch and its adjacent ones. This disparity is visually described by an inter-patch feature, where the hybrid effect of luminance masking and contrast masking is taken into account. The inter-patch similarity is further measured by modifying the normalized correlation coefficient (NCC). On the other hand, we also attach importance to the impact of image contents within one patch on the IQA problem. For the intra-patch feature, we consider image curvature as an important complement of image gradient. According to local image contents, the intra-patch similarity is measured by adaptively comparing image curvature and gradient. Besides, a nonlinear integration of the inter-patch and intra-patch similarity is presented to obtain an overall score of image quality. The experiments conducted on six publicly available image databases show that our scheme achieves better performance in comparison with several state-of-the-art schemes.

Generic properties of curvature sensing through vision and touch

Generic properties of curvature representations formed on the basis of vision and touch were examined as a function of mathematical properties of curved objects. Virtual representations of the curves were shown on a computer screen for visual scaling by sighted observers (experiment 1). Their physical counterparts were placed in the two hands of blindfolded and congenitally blind observers for tactile scaling. The psychophysical data show that curvature representations in congenitally blind individuals, who never had any visual experience, and in sighted observers, who rely on vision most of the time, are statistically linked to the same mathematical properties of the curves. The perceived magnitude of object curvature, sensed through either vision or touch, is related by a mathematical power law, with similar exponents for the two sensory modalities, to the aspect ratio of the curves, a scale invariant geometric property. This finding supports biologically motivated models of sensory integration suggesting a universal power law for the adaptive brain control and balance of motor responses to environmental stimuli from any sensory modality.

Quantification of changes in metamorphopsia and retinal contraction in eyes with spontaneous separation of idiopathic epiretinal membrane

BACKGROUND

To quantify changes in metamorphopsia and retinal contraction in eyes with idiopathic epiretinal membrane (ERM) before and after a spontaneous separation of ERM.

METHODS

Among 92 eyes of 92 patients with idiopathic ERM who were followed up at our hospital, 5 eyes of 5 patients had experienced a spontaneous separation of ERM during the follow-up period. Patient's metamorphopsia was assessed horizontally and vertically by a metamorphopsia chart developed by our group, M-CHARTS, to obtain the horizontal (MH) and vertical (MV) metamorphopsia scores. Difference in the scores before and after the membrane separation represents change in patient's metamorphopsia. Changes in retinal contraction were also evaluated horizontally and vertically with our original software for fundus image analysis. The difference between M-CHARTS scores and distances of retinal vessel movements with before and after membrane separation were measured.

RESULTS

All five subjects showed a decrease in the retinal contraction. Improved visual acuity was observed in three subjects, and no change was seen in the other two. Four subjects obtained better metamorphopsia scores after the membrane separation, while the other one was not detected with metamorphopsia by M-CHARTS either before or after the separation. In subjects with an improved MV, horizontal retinal movement was seen larger than the vertical movement. Similarly, the subjects with an improved MH indicated a larger vertical retinal movement than the horizontal movement.

CONCLUSIONS

The direction of patient's metamorphopsia closely associated with the direction of retinal contraction before and after a spontaneous separation of ERM.

“Phase capture” in amblyopia: The influence function for sampled shape

This study was concerned with what stimulus information humans with amblyopia use to judge the shape of simple objects. We used a string of four Gabor patches to define a contour. A fifth, center patch served as the test pattern. The observers' task was to judge the location of the test pattern relative to the contour. The contour was either a straight line, or an arc with positive or negative curvature. We asked whether phase shifts in the inner or outer pairs of patches distributed along the contour influence the perceived shape. That is, we measured the phase shift influence function. Our results, consistent with previous studies, show that amblyopes are imprecise in shape discrimination, showing elevated thresholds for both lines and curves. We found that amblyopes often make much larger perceptual errors (biases) than do normal observers in the absence of phase shifts. These errors tend to be largest for curved shapes and at large separations. In normal observers, shifting the phase of inner patches of the string by 0.25 cycle results in almost complete phase capture (attraction) at the smallest separation (2 lambda), and the capture effect falls off rapidly with separation. A 0.25 cycle shift of the outer pair of patches has a much smaller effect, in the opposite direction (repulsion). While several amblyopic observers showed reduced capture by the phase of the inner patches, to our surprise, several of the amblyopes were sensitive to the phase of the outer patches. We used linear multiple regression to determine the weights of all cues to the task: the carrier phase of the inner patches, carrier phase of the outer patches and the envelope of the outer patches. Compared to normal observers, some amblyopes show a weaker influence of the phase of the inner patches, and a stronger influence of both the phase and envelope of the outer patches. We speculate that this may be a consequence of abnormal "crowding" of the inner patches by the outer ones.

"Phase capture" in the perception of interpolated shape: cue combination and the influence function

This study was concerned with what stimulus information observers use to judge the shape of simple objects. We used a string of four Gabor patches to define a contour. A fifth, center patch served as a test pattern. The observers' task was to judge the location of the test pattern relative to the contour. The contour was either a straight line, or an arc with positive or negative curvature (the radius of curvature was either 2 or 6 deg). We asked whether phase shifts in the inner or outer pairs of patches distributed along the contour influence the perceived shape. That is, we measured the phase shift influence function. We found that shifting the inner patches of the string by 0.25 cycle results in almost complete phase capture (attraction) at the smallest separation (2 lambda), and the capture effect falls off rapidly with separation. A 0.25 cycle shift of the outer pair of patches has a much smaller effect, in the opposite direction (repulsion). In our experiments, the contour is defined by two cues--the cue provided by the Gabor carrier (the 'feature' cue) and that defined by the Gaussian envelope (the 'envelope' cue). Our phase shift influence function can be thought of as a cue combination task. An ideal observer would weight the cues by the inverse variance of the two cues. The variance in each of these cues predicts the main features of our results quite accurately.

Uniformity and asymmetry of rapid curved-line detection explained by parallel categorical coding of contour curvature

The aim of this work was to elucidate several characteristic phenomena associated with rapid curved-line detection in multi-element arrays and to provide a unified account of the underlying curvature-sensitive mechanisms. To this end, a parametric experiment was performed in which the detectability of a curved-line target in a briefly presented planar array of curved-line distractors was measured for a range of target and distractor curvatures and distractor numbers. For both vertically oriented and randomly oriented curved lines, it was found that (1) the dependence of target detectability on target curvature was independent of distractor number for small distractor curvatures but not for medium-to-large distractor curvatures; (2) an asymmetry in target detectability with respect to interchange of target and distractor curvatures occurred only with large distractor numbers; and (3) with small distractor numbers, target detectability depended only on the difference between target and distractor curvatures. These properties of spatial parallelism, asymmetry, and uniformity were explained quantitatively by a minimal model of rapid curved-line detection in which contour curvature was coded in terms of just two or three curvature categories, depending on curved-line orientation.

Detecting disorder in spatial vision

In normal foveal vision, visual space is accurately mapped from retina to cortex. However, the normal periphery, and the central field of strabismic amblyopes have elevated position discrimination thresholds, which have often been ascribed to increased 'intrinsic' spatial disorder. In the present study we evaluated the sensitivity of the human visual system (both normal and amblyopic) to spatial disorder, and asked whether there is increased 'intrinsic' topographical disorder in the amblyopic visual system. Specifically, we measured thresholds for detecting disorder (two-dimensional Gaussian position perturbations) either in a horizontal string of N equally spaced samples (Gabor patches), or in a ring of equally spaced samples over a wide range of feature separations. We also estimated both the 'equivalent intrinsic spatial disorder' and sampling efficiency using an equivalent noise approach. Our results suggest that both thresholds for detecting disorder, and equivalent intrinsic disorder depend strongly on separation, and are modestly increased in strabismic amblyopes. Strabismic amblyopes also show markedly reduced sampling efficiency. However, neither amblyopic nor peripheral vision performs like ideal or human observers with added separation-independent positional noise. Rather, the strong separation dependence suggests that the 'equivalent intrinsic disorder' may not reflect topographic disorder at all, but rather may reflect an abnormality in the amblyopes' Weber relationship.

Are judgements of circularity local or global?

We assessed, in a task where subjects had to detect smooth deviations from circularity, whether the underlying mechanisms were localised in space to the size of the individual perturbations or whether they computed global shape. By manipulating the phase, the number of cycles of modulation and the spatial arrangement of the perturbations we argue that although either aspect can be detected, performance is ultimately limited by a global shape detecting mechanism. We show that this global mechanism receives input from spatially coarse, crossed orientationally tuned filters whose peak position in orientation depends on the overall shape to be detected.

Geometric representation of the mechanisms underlying human curvature detection

Combined manipulation of blur, line length and contrast reveal two distinct processes involved in curvature detection. When line length is small relative to blur, thresholds are almost directly proportional to blur and independent of line length. When line length is large relative to blur thresholds are directly proportional to line length and independent of blur. The aspect ratio (line length/blur) of curved contours represents a scale-invariant metric which forms the decisive factor in determining curvature performance.

Spatial uncertainty and sampling efficiency in amblyopic position acuity

Spatial uncertainty and undersampling are two of the major hypotheses for the losses of amblyopic spatial vision. To test these two hypotheses, equivalent spatial uncertainty and spatial integration efficiency in spatial position judgments were quantified with a spatial perturbation paradigm. Specifically, three-line bisection thresholds were measured for the amblyopic eyes of two strabismic and two anisometropic amblyopes, and for normal controls. The horizontal stimulus lines comprised discrete dark dots distributed randomly around the mean line position according to a gaussian function. Line separation, the number of dots on each line (N), stimulus contrast (C), and the vertical standard deviation (sigma e) of the dot distribution were varied. An ideal observer analysis quantified the magnitude of equivalent spatial uncertainty (sigma s), the effective number of dots used (k), and spatial integration efficiency (k/N). At the optimal separation, equivalent spatial uncertainty (sigma s) is approximately ten-fold higher in both types of amblyopic visual systems than in control observers, even when stimulus visibility is accounted for. This apparent increase in sigma s is largely due to a shift in spatial scale of analysis in the amblyopic eye. Integration efficiency (k/N) increases in proportion to stimulus contrast or visibility (in units of detection threshold). Unlike sigma s, k/N is different between the two types of amblyopia. For the anisometropic observers, k/N is quantitatively similar to that of control observers. For the strabismic observers, on the other hand, k/N is reduced even after taking stimulus visibility into account. The decreased spatial integration efficiency in the strabismic visual system suggests that spatial undersampling may occur at a secondary stage of visual processing, beyond the detection stage.

Intrinsic uncertainty and integration efficiency in bisection acuity

A spatial perturbation paradigm was used to determine equivalent intrinsic uncertainty and spatial integration efficiency in bisection. Specifically, three-line bisection thresholds were measured in the fovea of four normal observers with stimulus lines comprised of discrete dark dots distributed randomly around the mean line position according to a Gaussian function. The standard deviation of the Gaussian distribution (sigma e), the number (N), and the strength (C) of the dots as well as line separation were varied. Bisection thresholds were modeled by an ideal integrator, from which the magnitude of equivalent internal uncertainty (sigma i), the equivalent effective number of dots (k), and equivalent integration efficiency (k/N) were quantified. At the 2 min arc separation, sigma i decreases (down to a few sec arc) as N and/or C increases. The effects of both N and C can be accounted for by the stimulus visibility (V, in multiples of detection threshold). At the 16 min arc separation, sigma i is independent of N, C, or V, and is about 1 min arc. The two different forms of sigma i indicate that bisection judgments are limited by at least two separate sources of limiting noise, consistent with the hypothesis of two separate mechanisms (i.e. spatial filters and local signs). A visibility dependent sigma i at the 2 min arc separation can be explained on the basis of contrast sensitive spatial filter mechanisms. A fixed sigma i at the 16 min arc separation indicates a genuine positional uncertainty, consistent with local-sign mechanisms. Interestingly, equivalent integration efficiency (k/N) is very similar at the two line separations. k/N is critically dependent on, and proportional to C, indicating a common limitation in a detection mechanism.

The perceived orientation of aliased lines

The use of raster display devices for the display of graphics causes problems of aliasing when edges or lines are produced. This can be significant in those psychophysical experiments where the orientational properties of the stimulus are important. We have assessed the perceived orientation of a selection of aliased lines by comparing them with the orientation of pairs of dots. It is found that the perceptual orientation is modelled well by a least-squares metric on the pixels that compose the line. Small deviations from this metric were found, and were also found in a control experiment employing anti-aliased lines. They appear to be due to range effects. Averaged across subjects, orientational acuity was only slightly lower for aliased lines.

Extrapolation of linear motion

We investigated observers' ability to extrapolate a linear trajectory of a moving point, in order to determine how effectively the visual system can combine orientation and position information for moving stimuli. Observers saw a probe dot moving along a straight line toward a stationary target dot. The probe dot extinguished before reaching the target, and the observers' task was to judge whether an extrapolation of the trajectory of the probe would pass to the left or right of the target. Performance was measured as a function of probe velocity, length of the visible trajectory, and location of the target. The empirical results indicated that over a range of conditions, performance on this task is qualitatively similar to, but somewhat less accurate than, that on an analogous task with static stimuli. A four-component model is presented to account for the results. The model specifies an accurate extraction of probe motion parameters, extrapolation of the motion by an ideal observer, and limitations on the input to these processes in the form of visual field spatial inhomogeneity and temporal decay of position information.

Regional distribution of the mechanisms that underlie spatial localization

In order to understand the regional distribution of the mechanisms which underlie localization accuracy we (1) chose a task which is known to involve localization accuracy (2) optimized stimulus parameters for ecentric loci and (3) determined how two key spatial factors which affect localization accuracy vary as a function of ecentricity. These involve Gaussian blur and Gaussian jitter. These results suggest that there are three different functions with ecentricity for the mechanisms underlying this task which we relate to the spatial properties of the retina, namely mean cone density, receptoral convergence and regularity.