High Spatial Frequency Images Research Articles

Scene content is predominantly conveyed by high spatial frequencies in scene-selective visual cortex.

In complex real-world scenes, image content is conveyed by a large collection of intertwined visual features. The visual system disentangles these features in order to extract information about image content. Here, we investigate the role of one integral component: the content of spatial frequencies in an image. Specifically, we measure the amount of image content carried by low versus high spatial frequencies for the representation of real-world scenes in scene-selective regions of human visual cortex. To this end, we attempted to decode scene categories from the brain activity patterns of participants viewing scene images that contained the full spatial frequency spectrum, only low spatial frequencies, or only high spatial frequencies, all carefully controlled for contrast and luminance. Contrary to the findings from numerous behavioral studies and computational models that have highlighted how low spatial frequencies preferentially encode image content, decoding of scene categories from the scene-selective brain regions, including the parahippocampal place area (PPA), was significantly more accurate for high than low spatial frequency images. In fact, decoding accuracy was just as high for high spatial frequency images as for images containing the full spatial frequency spectrum in scene-selective areas PPA, RSC, OPA and object selective area LOC. We also found an interesting dissociation between the posterior and anterior subdivisions of PPA: categories were decodable from both high and low spatial frequency scenes in posterior PPA but only from high spatial frequency scenes in anterior PPA; and spatial frequency was explicitly decodable from posterior but not anterior PPA. Our results are consistent with recent findings that line drawings, which consist almost entirely of high spatial frequencies, elicit a neural representation of scene categories that is equivalent to that of full-spectrum color photographs. Collectively, these findings demonstrate the importance of high spatial frequencies for conveying the content of complex real-world scenes.

Enhancing visual performance in individuals with cortical visual impairment (homonymous hemianopsia): Tapping into blindsight

Abstract Homonymous hemianopsia is a type of cortical blindness in which vision is lost completely or partially in the left half or the right half of the field of vision. It is prevalent in approximately 12% of traumatic brain injury and 35% of strokes. Patients often experience difficulty with activities such as ambulating, eating, reading, and driving. Due to the high prevalence of homonymous hemianopsia and its associated difficulties, it is imperative to find methods for visual rehabilitation in this condition. Traditional methods such as prism glasses can cause visual confusion and result in patient noncompliance. There is a large unmet medical need for improving this condition. In this article, we propose that modifying visual stimuli to activate non-cortical areas of visual processing, such as lateral geniculate nucleus and superior colliculus, may result in increased visual awareness. Presenting high contrast and low spatial frequency visual stimuli can increase visual detection in patients with cortical blindness, a phenomenon known as blindsight. Augmented virtual reality goggles have the potential to alter real-time visual input to high contrast and low spatial frequency images, possibly improving visual detection in the blind hemifield and providing an alternative therapy for homonymous hemianopsia.

Calibration Method for the Panel-type Multi-view Display

School of Electrical Engineering, Seoul National University, Gwanak-ro 1, Gwanak-gu, Seoul 151-744, Korea(Received April 24, 2015 : revised August 17, 2015 : accepted September 1, 2015)We propose a novel calibration method which can be applied to all kinds of panel-type multi-view displays. We analyze how the angular, the axial, and the lateral misalignment affects the 3D image quality in a panel-type multi-view display. We demonstrate the ray optics simulation with a 3-view slanted parallax barrier system using pentile display for the quantitative calculation. Based on the analysis, we propose a new alignment pattern for all kinds of panel-type multi-view displays. The proposed pattern is sensitive to all of the angular, the axial, and the lateral misalignments. The high spatial frequency images and on and off alignment in the proposed pattern help observers to calibrate the system easily. We theoretically show the generality of the proposed alignment pattern and verify the pattern with image simulations and experiments.Keywords : Multi-view display, 3D displayOCIS codes : (100.6890) Three-dimensional image processing; (110.2990) Image formation theory

Aberrant early visual neural activity and brain-behavior relationships in anorexia nervosa and body dysmorphic disorder.

Background: Body dysmorphic disorder (BDD) and anorexia nervosa (AN) share the clinical symptom of disturbed body image, which may be a function of perceptual distortions. Previous studies suggest visual or visuospatial processing abnormalities may be contributory, but have been unable to discern whether these occur early or late in the visual processing stream. We used electroencephalography (EEG) and visual event related potentials (ERP) to investigate early perceptual neural activity associated with processing visual stimuli.Methods: We performed EEG on 20 AN, 20 BDD, 20 healthy controls, all unmedicated. In order to probe configural/holistic and detailed processing, participants viewed photographs of faces and houses that were unaltered or filtered to low or high spatial frequencies, respectively. We calculated the early ERP components P100 and N170, and compared amplitudes and latencies among groups.Results: P100 amplitudes were smaller in AN than BDD and healthy controls, regardless of spatial frequency or stimulus type (faces or houses). Similarly, N170 latencies were longer in AN than healthy controls, regardless of spatial frequency or stimulus type, with a similar pattern in BDD at trend level significance. N170 amplitudes were smaller in AN than controls for high and normal spatial frequency images, and smaller in BDD than controls for normal spatial frequency images, regardless of stimulus type. Poor insight correlated with lower N170 amplitudes for normal and low spatial frequency faces in the BDD group.Conclusions: Individuals with AN exhibit abnormal early visual system activity, consistent with reduced configural processing and enhanced detailed processing. This is evident regardless of whether the stimuli are appearance–or non-appearance-related, and thus may be a reflection of general, early perceptual abnormalities. As N170 amplitude could be a marker of structural encoding of faces, lower values may be associated with perceptual distortions and could contribute to poor insight in BDD. Future studies may explore visual ERP measures as potential biomarkers of illness phenotype.

Early and late effects of objecthood and spatial frequency on event-related potentials and gamma band activity.

BackgroundThe visual system may process spatial frequency information in a low-to-high, coarse-to-fine sequence. In particular, low and high spatial frequency information may be processed via different pathways during object recognition, with LSF information projected rapidly to frontal areas and HSF processed later in visual ventral areas. In an electroencephalographic study, we examined the time course of information processing for images filtered to contain different ranges of spatial frequencies. Participants viewed either high spatial frequency (HSF), low spatial frequency (LSF), or unfiltered, broadband (BB) images of objects or non-object textures, classifying them as showing either man-made or natural objects, or non-objects. Event-related potentials (ERPs) and evoked and total gamma band activity (eGBA and tGBA) recorded using the electroencephalogram were compared for object and non-object images across the different spatial frequency ranges.ResultsThe visual P1 showed independent modulations by object and spatial frequency, while for the N1 these factors interacted. The P1 showed more positive amplitudes for objects than non-objects, and more positive amplitudes for BB than for HSF images, which in turn evoked more positive amplitudes than LSF images. The peak-to-peak N1 showed that the N1 was much reduced for BB non-objects relative to all other images, while HSF and LSF non-objects still elicited as negative an N1 as objects. In contrast, eGBA was influenced by spatial frequency and not objecthood, while tGBA showed a stronger response to objects than non-objects.ConclusionsDifferent pathways are involved in the processing of low and high spatial frequencies during object recognition, as reflected in interactions between objecthood and spatial frequency in the visual N1 component. Total gamma band seems to be related to a late, probably high-level representational process.

The highs and lows of object impossibility: effects of spatial frequency on holistic processing of impossible objects.

Holistic processing, the decoding of a stimulus as a unified whole, is a basic characteristic of object perception. Recent research using Garner's speeded classification task has shown that this processing style is utilized even for impossible objects that contain an inherent spatial ambiguity. In particular, similar Garner interference effects were found for possible and impossible objects, indicating similar holistic processing styles for the two object categories. In the present study, we further investigated the perceptual mechanisms that mediate such holistic representation of impossible objects. We relied on the notion that, whereas information embedded in the high-spatial-frequency (HSF) content supports fine-detailed processing of object features, the information conveyed by low spatial frequencies (LSF) is more crucial for the emergence of a holistic shape representation. To test the effects of image frequency on the holistic processing of impossible objects, participants performed the Garner speeded classification task on images of possible and impossible cubes filtered for their LSF and HSF information. For images containing only LSF, similar interference effects were observed for possible and impossible objects, indicating that the two object categories were processed in a holistic manner. In contrast, for the HSF images, Garner interference was obtained only for possible, but not for impossible objects. Importantly, we provided evidence to show that this effect could not be attributed to a lack of sensitivity to object possibility in the LSF images. Particularly, even for full-spectrum images, Garner interference was still observed for both possible and impossible objects. Additionally, performance in an object classification task revealed high sensitivity to object possibility, even for LSF images. Taken together, these findings suggest that the visual system can tolerate the spatial ambiguity typical to impossible objects by relying on information embedded in LSF, whereas HSF information may underlie the visual system's susceptibility to distortions in objects' spatial layouts.

Temporal dynamics of spatial frequency processing in infants.

The current study examined the temporal dynamics of coarse and fine spatial information processing in 7- to 8-month-old infants. The ability to discriminate between spatially filtered images was assessed by measuring infants' spontaneous preference for a changing over no-changing image sequences. In Experiments 1 and 2, we found that infants were able to discriminate between low spatial frequency (LSF) image sequences at shorter durations (150 ms) than was the case with high spatial frequency (HSF) images (300 ms). When the LSF and HSF changes were pitted against each other in hybrid images containing both spatial frequencies (Experiment 3), the 7- to 8-month-old infants showed a preference for the LSF change across all tested durations (150 ms to 600 ms). However, infants' processing of hybrid image sequences was modulated both by changes in the relative contrast energy between LSFs and HSFs (Experiment 4), and image duration (Experiment 5). Finally, we found that in 12- to 13-month-old infants, the shift toward HSF dominance occurred at shorter duration than in 7- to 8-month-old infants (Experiment 6). Our findings are among the first to provide a temporal characterization of coarse-to-fine processing in infants' perception. Possible links to the development of specialized visual pathways are discussed.

Hemi‐field and full‐field form‐deprivation induce timing changes in multifocalERGresponses in chick

In animal models hemi-field deprivation results in localised, graded vitreous chamber elongation and presumably deprivation induced localised changes in retinal processing. The aim of this research was to determine if there are variations in ERG responses across the retina in normal chick eyes and to examine the effect of hemi-field and full-field deprivation on ERG responses across the retina and at earlier times than have previously been examined electrophysiologically. Chicks were either untreated, wore monocular full-diffusers or half-diffusers (depriving nasal retina) (n = 6-8 each group) from day 8. mfERG responses were measured using the VERIS mfERG system across the central 18.2º× 16.7º (H × V) field. The stimulus consisted of 61 unscaled hexagons with each hexagon modulated between black and white according to a pseudorandom binary m-sequence. The mfERG was measured on day 12 in untreated chicks, following 4 days of hemi-field diffuser wear, and 2, 48 and 96 h after application of full-field diffusers. The ERG response of untreated chick eyes did not vary across the measured field; there was no effect of retinal location on the N1-P1 amplitude (p = 0.108) or on P1 implicit time (p > 0.05). This finding is consistent with retinal ganglion cell density of the chick varying by only a factor of two across the entire retina. Half-diffusers produced a ramped retina and a graded effect of negative lens correction (p < 0.0001); changes in retinal processing were localized. The untreated retina showed increasing complexity of the ERG waveform with development; form-deprivation prevented the increasing complexity of the response at the 2, 48 and 96 h measurement times and produced alterations in response timing. Form-deprivation and its concomitant loss of image contrast and high spatial frequency images prevented development of the ERG responses, consistent with a disruption of development of retinal feedback systems. The characterisation of ERG responses in normal and deprived chick eyes across the retina allows the assessment of concurrent visual and retinal manipulations in this model.

How action influences object perception

Although object perception is typically associated with the parvocellular (P) pathway, a form of fast “gist” object perception may be due to activity in the magnocellular (M) pathway (Kveraga et al., 2007). Because the M-pathway is typically associated with action, we hypothesized that manipulations of action would influence speeded object perception. In three experiments, participants indicated whether the objects shown in low and high spatial frequency (HSF) images were larger or smaller than a prototypical shoebox. In Experiment 1, participants used a proximal (hands on monitor) or distal (hands on keyboard) hand posture in separate blocks. In Experiment 2, only the proximal hand posture was used, but the hands were either action oriented with palms in (palms toward the stimuli) or non-action oriented with palms out (palms away from the stimuli). In Experiment 3, we used the palms-in proximal hand posture but manipulated the type of visual stimuli such that they were either action oriented (easily grasped) or non-action oriented (not easily grasped). In all three experiments, the advantage in identifying the low spatial frequency (LSF) images was greater when action was primed (proximal hands, palms-in, graspable). Together, these experiments show that the M-pathway is involved in rapid “gist” object perception, and this type of object perception is influenced by action.

The effect of spatial frequency and face inversion on facial expression processing in children with autism spectrum disorder

AbstractTo investigate whether facial expression processing in children with autism spectrum disorder (ASD) is based on local information of the stimuli, we prepared low spatial frequency (LSF) images with blurred facial features and high spatial frequency (HSF) images with rich facial features from broad (normal) spatial frequency (BSF) images. Eighteen children with ASD (mean age 11.9 years) and 19 typically developing (TD) children (mean age 11.4 years) matched on nonverbal IQ were presented these stimuli in upright and inverted orientations. The children with ASD had difficulty in processing facial expressions from the BSF and LSF images, but not from the HSF images. In addition, the BSF and HSF images elicited the inversion effect in the TD children, but not in the children with ASD. In contrast, the LSF images elicited the inversion effect in both groups of children. These results suggest that children with ASD are biased towards processing facial expression based on local information, even though their capacity to process facial expressions configurally is spared.

Differential Spatial Frequency Modulations on Facial Emotion Detection and Perception

The recognition of facial affect is a core component of social cognition and is influenced by bottom-up and top- down processes. Spatial frequency-specific processing is a fundamental mechanism underpinning visual perception that also contributes to emotion processing. Yet, it remains relatively unclear how this bottom-up process modulates analysis of affect signals. In this study, face images were manipulated via high-pass and low-pass spatial frequency filtering, and the effects of these manipulations were measured on two tasks: emotion detection and emotion perception. The detection task measured individuals' ability to detect the presence of happiness or fear when relying only upon high spatial frequency (HSF), low spatial frequency (LSF) or unfiltered (BSF) images. The perception task evaluated individuals' tendency to perceive one type of images (e.g. LSF) as happier or more fearful than another (e.g. HSF). On the detection task, individuals were better at detecting happiness in LSF images as compared to HSF images, a result not found in the fear detection condition. On the perception task, images with HSFs were perceived as significantly happier than images without HSFs, but only at low emotional salience levels. No significant effect for spatial frequency manipulation was found in the fear perception condition. Together, these results highlight the significant bottom-up role of spatial frequency content in perceiving happiness as opposed to fear and indicates the unique roles of bottom-up visual mechanisms when processing these two emotions.

Perceptual Learning of Inverted Faces across Different Spatial Frequency Bands

Practicing a perceptual task can improve the ability to detect, discriminate, and identify stimuli, a phenomenon known as perceptual learning (1). Improvements have been found for simple and complex visual stimuli, including upright faces (2,3). However, little is known about the effects of practice on recognition of inverted faces (4,5). Here we hypothesized that adults might be better at learning to recognize inverted faces if trained with low spatial frequency images (<5 cpi; LSF) than full spectrum (FULL) or high spatial frequency images (>24 cpi; HSF) since newborns learn to recognize upright faces despite poor visual acuity (6). We trained 3 groups of 16 adults in a 10-AFC task with LSF, HSF or FULL inverted faces presented with different viewpoints. Participants were exposed either to the same or a different set of faces on Day 2. Although reaction times became significantly faster for all groups during training, accuracy improved only in the FULL (14%) and HSF (7%) groups. In contrast, the LSF group's overall improvement was only 1% (Fig1). Improvements from before to after training were assessed by comparisons to an untrained control group on 4 tasks. The FULL group showed more improvement in both accuracy and reactions time than controls on the delayed matching task of full spectrum inverted faces, thereby demonstrating generalization of training to novel faces. They were also faster on the same task with upright faces. The HSF group showed increased and decreased holistic processing of full spectrum inverted and upright faces, respectively, as indexed by reaction times on the composite face effect. Despite no improvement in accuracy during training, the LSF group improved more than controls on reaction times to match simultaneously presented full-spectrum inverted faces. These results suggest that perceptual learning of inverted faces and its generalization depend strongly on their spatial frequency content.

Minimum average colour space entropy – colour component transform based colour image compression

The standards given by joint photographic expert group (JPEG and JPEG2000) utilise the component colour space transform (CCT) and subband transform (SBT) for the purpose of image compression. The CCT reduces the redundancy between the RGB components and relates the different components of image at every pixel position. On the other side, SBT relates the different image pixels with its surroundings or vicinity within a component. The minimum average colour space entropy (MACSE) based CCT further reduces the inter colour component redundancy by exploiting the distribution of image pixels in spatial domain. This is a reversible process. The MACSE-CCT along with JPEG coding shows around 1.5 dB improvement in peak signal to noise ratio for very low bit-rates. The performance of this hybrid model is found better for very high spatial frequency images from the point of view of visual perception.

Effects of spatial frequency bands on perceptual decision: It is not the stimuli but the comparison

Observers performed three between- and two within-category perceptual decisions with hybrid stimuli comprising low and high spatial frequency (SF) images. We manipulated (a) attention to, and (b) congruency of information in the two SF bands. Processing difficulty of the different SF bands varied across different categorization tasks: house-flower, face-house, and valence decisions were easier when based on high SF bands, while flower-face and gender categorizations were easier when based on low SF bands. Larger interference also arose from response relevant distracters that were presented in the "preferred" SF range of the task. Low SF effects were facilitated by short exposure durations. The results demonstrate that decisions are affected by an interaction of task and SF range and that the information from the non-attended SF range interfered at the decision level. A further analysis revealed that overall differences in the statistics of image features, in particular differences of orientation information between two categories, were associated with decision difficulty. We concluded that the advantage of using information from one SF range over another depends on the specific task requirements that built on the differences of the statistical properties between the compared categories.

PW01-173 - Object visual processing in body dysmorphic disorder

IntroductionBody dysmorphic disorder (BDD) is characterized by a preoccupation with misperceived defects in appearance. Neuropsychological studies and neuroimaging studies of face perception suggests visual processing abnormalities in individuals with BDD, which may involve focus on details at the expense of configural elements.ObjectiveThe objective of the current functional magnetic resonance imaging (fMRI) study was to determine whether individuals with BDD have abnormal patterns of brain activation when visually processing non-face objects (houses).MethodsFourteen medication-free subjects with BDD and fourteen matched healthy controls engaged in a matching task of photographs of houses that werea)unaltered,b)high spatial frequency (high detail elements only), orc)low spatial frequency (low detail elements only).The main outcome measure was group differences in blood oxygen level-dependent fMRI signal changes.ResultsBDD subjects demonstrated lesser activity relative to controls in left parahippocampal gyrus, left lingual gyrus, and bilateral precuneus for low spatial frequency images and relatively greater activity in the frontal pole, left superior frontal gyrus, bilateral anterior cingulate gyrus, and bilateral paracingulate gyrus for high spatial frequency images (Figure 1).ConclusionsIndividuals with BDD have abnormal brain activation patterns when viewing houses for both detailed and configural/holistic visual elements. These results suggest general abnormalities in higher- and lower-order visual processing in individuals with BDD, beyond that for appearance-related stimuli.

Nonlinear Amygdala Response to Face Trustworthiness: Contributions of High and Low Spatial Frequency Information

Previous neuroimaging research has shown amygdala sensitivity to the perceived trustworthiness of neutral faces, with greater responses to untrustworthy compared with trustworthy faces. This observation is consistent with the common view that the amygdala encodes fear and is preferentially responsive to negative stimuli. However, some studies have shown greater amygdala activation to positive compared with neutral stimuli. The first goal of this study was to more fully characterize the amygdala response to face trustworthiness by modeling its activation with both linear and nonlinear predictors. Using fMRI, we report a nonmonotonic response profile, such that the amygdala responds strongest to highly trustworthy and highly untrustworthy faces. This finding complicates future attempts to make inferences about mental states based on activation in the amygdala. The second goal of the study was to test for modulatory effects of image spatial frequency filtering on the amygdala response. We predicted greater amygdala sensitivity to face trustworthiness for low spatial frequency images compared with high spatial frequency images. Instead, we found that both frequency ranges provided sufficient information for the amygdala to differentiate faces on trustworthiness. This finding is consistent with behavioral results and suggests that trustworthiness information may reach the amygdala through pathways carrying both coarse and fine resolution visual signals.

The importance of low spatial frequency information for recognising fearful facial expressions

A recent brain imaging study (Vuilleumier, Armony, Driver and Dolan 2003, Nature Neuroscience, 6, 624–631) has shown that amygdala responses to fearful expressions are preferentially driven by intact or low spatial frequency (LSF) images of faces, rather than by high spatial frequency (HSF) images. These results suggest that LSF components processed rapidly via magnocellular pathways within the visual system might be very efficiently conveyed to the amygdala for the rapid recognition of fearful expressions, perhaps via a subcortical pathway that activates the pulvinar and superior colliculus, but which bypasses any finer visual analysis of HSF cues in the striate and temporal extrastriate cortex. The purpose of this paper is to analyse the statistical properties of LSF compared with HSF and intact faces. The statistical analysis shows that the LSF components in faces, which are typically extracted rapidly by the visual system, provide a better source of information than HSF components for the correct categorisation of fearful expressions in faces. These results support the idea that visual pathways from the magnocellular visual neurons might be optimal, at a computational level, for the rapid classification of fearful emotional expressions in human faces.

Quantitative X-ray projection microscopy: phase-contrast and multi-spectral imaging.

We outline a new approach to X-ray projection microscopy in a scanning electron microscope (SEM), which exploits phase contrast to boost the quality and information content of images. These developments have been made possible by the combination of a high-brightness field-emission gun (FEG)-based SEM, direct detection CCD technology and new phase retrieval algorithms. Using this approach we have been able to obtain spatial resolution of < 0.2 micro m and have demonstrated novel features such as: (i) phase-contrast enhanced visibility of high spatial frequency image features (e.g. edges and boundaries) over a wide energy range; (ii) energy-resolved imaging to simultaneously produce multiple quasi-monochromatic images using broad-band polychromatic illumination; (iii) easy implementation of microtomography; (iv) rapid and robust phase/amplitude-retrieval algorithms to enable new real-time and quantitative modes of microscopic imaging. These algorithms can also be applied successfully to recover object-plane information from intermediate-field images, unlocking the potentially greater contrast and resolution of the intermediate-field regime. Widespread applications are envisaged for fields such as materials science, biological and biomedical research and microelectronics device inspection. Some illustrative examples are presented. The quantitative methods described here are also very relevant to projection microscopy using other sources of radiation, such as visible light and electrons.

Is It Warm? Is It Real? Or Just Low Spatial Frequency?

Leonardo da Vinci's portrait of the Mona Lisa is famous for her smile. Perhaps it is the difference in her expression carried by high and low spatial frequency ranges (gradual versus sharp luminance gradations) that helps produce her smile's elusive quality. The spatial resolution of the human visual system changes dramatically with distance from the center of gaze ([1][1]), due to the fact that both the retina and the visual cortex devote disproportionately more neuronal machinery to the fovea. Acuity 6 to 7° eccentric of the center of gaze is about one-tenth the acuity at the center of gaze. This means that our central vision is dominated by significantly higher spatial frequencies than is our peripheral vision. Conversely, vision only a few degrees from the center of gaze is much blurrier than in the fovea. To see how Mona Lisa's smile would look at different eccentricities, the image has been filtered to exaggerate selectively low or high spatial frequencies. A clear smile is much more apparent in the low spatial frequency images than in the high spatial frequency image. Thus, if you look at the painting so that your gaze falls on the background or on Mona Lisa's hands, your perception of her mouth would be dominated by low spatial frequencies, so it would appear much more cheerful than when you look directly at her mouth. This explanation goes beyond the popular idea that da Vinci blurred her mouth ( sfumato ) to make her expression ambiguous ([2][2]). It seems that her smile is more apparent in the low spatial frequency range, and therefore more apparent to peripheral vision than to central vision. Hence the elusive quality—you can't catch her smile by looking at her mouth. She smiles until you look at her mouth, and then it fades, like a dim star that disappears when you look directly at it. 1. [↵][3]1. T. Wertheim , Zeit. fur Psychol. Physiol. Sinnersorgen. 7, 172 (1894). [OpenUrl][4] 2. [↵][5]1. E. H. Gombrich , The Story of Art (Phaidon, London, 1999), p. 300. [1]: #ref-1 [2]: #ref-2 [3]: #xref-ref-1-1 View reference 1 in text [4]: {openurl}?query=rft.jtitle%253DZeit.%2Bf%25FCr%2BPsychol.%2BPhysiol.%2BSinnersorgen.%26rft.volume%253D7%26rft.spage%253D172%26rft.atitle%253DZEIT%2BFR%2BPSYCHOL%2BPHYSIOL%2BSINNERSORGEN%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [5]: #xref-ref-2-1 View reference 2 in text

High-performance autofocus circuit for biological microscopy

An autofocus circuit, based on measurement of the high spatial frequency image components, was designed for automated microscopic scanning of biological specimens. By careful consideration of the system transfer function, elimination of video end-of-line filter artifacts, correction for illumination instability, and incorporation of autogain, the focus measurement circuit attained the sensitivity and dynamic range necessary for robust operation even at the extremes of biological specimen detail encountered in exhaustive raster scans of thousands of fields. This new circuit exhibited a 25-fold improvement in dynamic range over a previous analog implementation, matched real-time digital performance at an order of magnitude lower cost, resulted in autofocus precision of 56 nm (or 10-fold better than the depth of field of the objective) in scanning experiments comprising over 10 000 microscope fields, and tracked focus at scanning speeds of up to 3.45 fields/s. Focus was correctly maintained in these scanning experiments without additional compensation for low-detail images. This circuit makes possible the widespread inclusion of high-performance autofocus as a low cost option in video microscopy systems.