Cortical Surround Interactions and Perceptual Salience via Natural Scene Statistics

Spatial context in images induces perceptual phenomena associated with salience and modulates the responses of neurons in primary visual cortex (V1). However, the computational and ecological principles underlying contextual effects are incompletely understood. We introduce a model of natural images that includes grouping and segmentation of neighboring features based on their joint statistics, and we interpret the firing rates of V1 neurons as performing optimal recognition in this model. We show that this leads to a substantial generalization of divisive normalization, a computation that is ubiquitous in many neural areas and systems. A main novelty in our model is that the influence of the context on a target stimulus is determined by their degree of statistical dependence. We optimized the parameters of the model on natural image patches, and then simulated neural and perceptual responses on stimuli used in classical experiments. The model reproduces some rich and complex response patterns observed in V1, such as the contrast dependence, orientation tuning and spatial asymmetry of surround suppression, while also allowing for surround facilitation under conditions of weak stimulation. It also mimics the perceptual salience produced by simple displays, and leads to readily testable predictions. Our results provide a principled account of orientation-based contextual modulation in early vision and its sensitivity to the homogeneity and spatial arrangement of inputs, and lends statistical support to the theory that V1 computes visual salience.     

Analysis of machine-selected cells with an image analysis system in normal and abnormal cervical specimens

An analysis has been performed of visual diagnostic criteria used in cervical cytology applied to machine selected cells in relation to automated classification based on variables, which can be recorded in an image system with automated cell search and segmentation, feature extraction and classification. A 98% accuracy could be obtained with the choice of the most ideal statistical methods for discrimination and the use of the most powerful variables recorded in the image system when compared with consensus of the visual diagnoses based on established cytological criteria for diagnosis of cancer and precancer of the cervix uteri. The most powerful discriminatory variables in the image system (of 17 recorded) for discrimination between normal and abnormal epithelial cells were, in addition to nuclear extinction, cytoplasmic extinction and cytoplasmic shape. It is concluded that the visual classification of cervical cells is highly accurate with experienced observers and that imaging microscopes can be trained to nearly equal this accuracy with appropriate statistical methods of discrimination. The problem of creating fully automated systems, however, also requires the inclusion of even more effective discriminatory variables and also the solution of such problems as automatic cell search, segmentation, artifact rejection, feature extraction, classification and electronic stability in order to become cost-effective.     

Psychophysical tests of the hypothesis of a bottom-up saliency map in primary visual cortex

A unique vertical bar among horizontal bars is salient and pops out perceptually. Physiological data have suggested that mechanisms in the primary visual cortex (V1) contribute to the high saliency of such a unique basic feature, but indicated little regarding whether V1 plays an essential or peripheral role in input-driven or bottom-up saliency. Meanwhile, a biologically based V1 model has suggested that V1 mechanisms can also explain bottom-up saliencies beyond the pop-out of basic features, such as the low saliency of a unique conjunction feature such as a red vertical bar among red horizontal and green vertical bars, under the hypothesis that the bottom-up saliency at any location is signaled by the activity of the most active cell responding to it regardless of the cell's preferred features such as color and orientation. The model can account for phenomena such as the difficulties in conjunction feature search, asymmetries in visual search, and how background irregularities affect ease of search. In this paper, we report nontrivial predictions from the V1 saliency hypothesis, and their psychophysical tests and confirmations. The prediction that most clearly distinguishes the V1 saliency hypothesis from other models is that task-irrelevant features could interfere in visual search or segmentation tasks which rely significantly on bottom-up saliency. For instance, irrelevant colors can interfere in an orientation-based task, and the presence of horizontal and vertical bars can impair performance in a task based on oblique bars. Furthermore, properties of the intracortical interactions and neural selectivities in V1 predict specific emergent phenomena associated with visual grouping. Our findings support the idea that a bottom-up saliency map can be at a lower visual area than traditionally expected, with implications for top-down selection mechanisms.     

Pooling strategies in V1 can account for the functional and structural diversity across species

Neurons in the primary visual cortex are selective to orientation with various degrees of selectivity to the spatial phase, from high selectivity in simple cells to low selectivity in complex cells. Various computational models have suggested a possible link between the presence of phase invariant cells and the existence of orientation maps in higher mammals’ V1. These models, however, do not explain the emergence of complex cells in animals that do not show orientation maps. In this study, we build a theoretical model based on a convolutional network called Sparse Deep Predictive Coding (SDPC) and show that a single computational mechanism, pooling, allows the SDPC model to account for the emergence in V1 of complex cells with or without that of orientation maps, as observed in distinct species of mammals. In particular, we observed that pooling in the feature space is directly related to the orientation map formation while pooling in the retinotopic space is responsible for the emergence of a complex cells population. Introducing different forms of pooling in a predictive model of early visual processing as implemented in SDPC can therefore be viewed as a theoretical framework that explains the diversity of structural and functional phenomena observed in V1.     

Independent Component Analysis of Temporal Sequences Subject to Constraints by Lateral Geniculate Nucleus Inputs Yields All the Three Major Cell Types of the Primary Visual Cortex

Information maximization has long been suggested as the underlying coding strategy of the primary visual cortex (V1). Grouping image sequences into blocks has been shown by others to improve agreement between experiments and theory. We have studied the effect of temporal convolution on the formation of spatiotemporal filters—that is, the analogues of receptive fields—since this temporal feature is characteristic to the response function of lagged and nonlagged cells of the lateral geniculate nucleus. Concatenated input sequences were used to learn the linear transformation that maximizes the information transfer. Learning was accomplished by means of principal component analysis and independent component analysis. Properties of the emerging spatiotemporal filters closely resemble the three major types of V1 cells: simple cells with separable receptive field, simple cells with nonseparable receptive field, and complex cells.     

Learning receptive field properties of complex cells in V1

There are two distinct classes of cells in the primary visual cortex (V1): simple cells and complex cells. One defining feature of complex cells is their spatial phase invariance; they respond strongly to oriented grating stimuli with a preferred orientation but with a wide range of spatial phases. A classical model of complete spatial phase invariance in complex cells is the energy model, in which the responses are the sum of the squared outputs of two linear spatially phase-shifted filters. However, recent experimental studies have shown that complex cells have a diverse range of spatial phase invariance and only a subset can be characterized by the energy model. While several models have been proposed to explain how complex cells could learn to be selective to orientation but invariant to spatial phase, most existing models overlook many biologically important details. We propose a biologically plausible model for complex cells that learns to pool inputs from simple cells based on the presentation of natural scene stimuli. The model is a three-layer network with rate-based neurons that describes the activities of LGN cells (layer 1), V1 simple cells (layer 2), and V1 complex cells (layer 3). The first two layers implement a recently proposed simple cell model that is biologically plausible and accounts for many experimental phenomena. The neural dynamics of the complex cells is modeled as the integration of simple cells inputs along with response normalization. Connections between LGN and simple cells are learned using Hebbian and anti-Hebbian plasticity. Connections between simple and complex cells are learned using a modified version of the Bienenstock, Cooper, and Munro (BCM) rule. Our results demonstrate that the learning rule can describe a diversity of complex cells, similar to those observed experimentally.     

A computational model to link psychophysics and cortical cell activation patterns in human texture processing

The human visual system uses texture information to automatically, or pre-attentively, segregate parts of the visual scene. We investigate the neural substrate underlying human texture processing using a computational model that consists of a hierarchy of bi-directionally linked model areas. The model builds upon two key hypotheses, namely that (i) texture segregation is based on boundary detection--rather than clustering of homogeneous items--and (ii) texture boundaries are detected mainly on the basis of a large scenic context that is analyzed by higher cortical areas within the ventral visual pathway, such as area V4. Here, we focus on the interpretation of key results from psychophysical studies on human texture segmentation. In psychophysical studies, texture patterns were varied along several feature dimensions to systematically characterize human performance. We use simulations to demonstrate that the activation patterns of our model directly correlate with the psychophysical results. This allows us to identify the putative neural mechanisms and cortical key areas which underlie human behavior. In particular, we investigate (i) the effects of varying texture density on target saliency, and the impact of (ii) element alignment and (iii) orientation noise on the detectability of a pop-out bar. As a result, we demonstrate that the dependency of target saliency on texture density is linked to a putative receptive field organization of orientation-selective neurons in V4. The effect of texture element alignment is related to grouping mechanisms in early visual areas. Finally, the modulation of cell activity by feedback activation from higher model areas, interacting with mechanisms of intra-areal center-surround competition, is shown to result in the specific suppression of noise-related cell activities and to improve the overall model capabilities in texture segmentation. In particular, feedback interaction is crucial to raise the model performance to the level of human observers.     

The dynamics of visual adaptation to faces

Several recent demonstrations using visual adaptation have revealed high-level aftereffects for complex patterns including faces. While traditional aftereffects involve perceptual distortion of simple attributes such as orientation or colour that are processed early in the visual cortical hierarchy, face adaptation affects perceived identity and expression, which are thought to be products of higher-order processing. And, unlike most simple aftereffects, those involving faces are robust to changes in scale, position and orientation between the adapting and test stimuli. These differences raise the question of how closely related face aftereffects are to traditional ones. Little is known about the build-up and decay of the face aftereffect, and the similarity of these dynamic processes to traditional aftereffects might provide insight into this relationship. We examined the effect of varying the duration of both the adapting and test stimuli on the magnitude of perceived distortions in face identity. We found that, just as with traditional aftereffects, the identity aftereffect grew logarithmically stronger as a function of adaptation time and exponentially weaker as a function of test duration. Even the subtle aspects of these dynamics, such as the power-law relationship between the adapting and test durations, closely resembled that of other aftereffects. These results were obtained with two different sets of face stimuli that differed greatly in their low-level properties. We postulate that the mechanisms governing these shared dynamics may be dissociable from the responses of feature-selective neurons in the early visual cortex.     

A Feedback Model of Attention Explains the Diverse Effects of Attention on Neural Firing Rates and Receptive Field Structure

Visual attention has many effects on neural responses, producing complex changes in firing rates, as well as modifying the structure and size of receptive fields, both in topological and feature space. Several existing models of attention suggest that these effects arise from selective modulation of neural inputs. However, anatomical and physiological observations suggest that attentional modulation targets higher levels of the visual system (such as V4 or MT) rather than input areas (such as V1). Here we propose a simple mechanism that explains how a top-down attentional modulation, falling on higher visual areas, can produce the observed effects of attention on neural responses. Our model requires only the existence of modulatory feedback connections between areas, and short-range lateral inhibition within each area. Feedback connections redistribute the top-down modulation to lower areas, which in turn alters the inputs of other higher-area cells, including those that did not receive the initial modulation. This produces firing rate modulations and receptive field shifts. Simultaneously, short-range lateral inhibition between neighboring cells produce competitive effects that are automatically scaled to receptive field size in any given area. Our model reproduces the observed attentional effects on response rates (response gain, input gain, biased competition automatically scaled to receptive field size) and receptive field structure (shifts and resizing of receptive fields both spatially and in complex feature space), without modifying model parameters. Our model also makes the novel prediction that attentional effects on response curves should shift from response gain to contrast gain as the spatial focus of attention drifts away from the studied cell.     

Anatomical Feature Segmentation of Femur Point Cloud Based on Medical Semantics

Feature segmentation is an essential phase for geometric modeling and shape processing in anatomical study of human skeleton and clinical digital treatment of orthopedics. Due to various degrees of freedom of bone surface, the existing segmentation algorithms can hardly meet specific medical need. To address this, a novel segmentation methodology for anatomical features of femur model based on medical semantics is put forward. First, anatomical reference objects (ARO) are created to represent typical characteristics of femur anatomy by 3D point fitting in combination with medical priori knowledge. Then, local point clouds between adjacent anatomies are selected according to the AROs to extract boundary feature point (BFP)s. Finally, the complete model of femur is divided into anatomical regions by executing the enhanced watershed algorithm guided with BFPs. Experimental results show that the proposed method has the advantages of automatic segmentation of femoral head, neck and other complex areas, and the segmentation results have better medical semantics. In addition, the slight modification of segmentation results can be achieved by adjusting a few threshold parameter values, which improves the convenience of modification for ordinary users.     

Adult visual cortical plasticity

The visual cortex has the capacity for experience-dependent change, or cortical plasticity, that is retained throughout life. Plasticity is invoked for encoding information during perceptual learning, by internally representing the regularities of the visual environment, which is useful for facilitating intermediate-level vision-contour integration and surface segmentation. The same mechanisms have adaptive value for functional recovery after CNS damage, such as that associated with stroke or neurodegenerative disease. A common feature to plasticity in primary visual cortex (V1) is an association field that links contour elements across the visual field. The circuitry underlying the association field includes a plexus of long-range horizontal connections formed by cortical pyramidal cells. These connections undergo rapid and exuberant sprouting and pruning in response to removal of sensory input, which can account for the topographic reorganization following retinal lesions. Similar alterations in cortical circuitry may be involved in perceptual learning, and the changes observed in V1 may be representative of how learned information is encoded throughout the cerebral cortex.     

Neural mechanisms of human texture processing: texture boundary detection and visual search

Texture of various appearances, geometric distortions, spatial frequency content and densities is utilized by the human visual system to segregate items from background and to enable recognition of complex geometric forms. For automatic, or pre-attentive, segmentation of a visual scene, sophisticated analysis and comparison of surface properties over wide areas of the visual field are required. We investigated the neural substrate underlying human texture processing, particularly the computational mechanisms of texture boundary detection. We present a neural network model which uses as building blocks model cortical areas that are bi-directionally linked to implement cycles of feedforward and feedback interaction for signal detection, hypothesis generation and testing within the infero-temporal pathway of form processing. In the spirit of Jake Beck's early investigations our model particularly builds upon two key hypotheses, namely that (i) texture segregation is based on boundary detection, rather than clustering homogeneous items, and (ii) texture boundaries are detected mainly on the basis of larger scenic contexts mediated by higher cortical areas, such as area V4. The latter constraint provides a basis for element grouping in accordance to the Gestalt laws of similarity and good continuation. It is shown through simulations that the model integrates a variety of psychophysical findings on texture processing and provides a link to the underlying physiology. The functional role of feedback processing is demonstrated by context dependent modulation of V1 cell activation, leading to sharply localized detection of texture boundaries. It furthermore explains why pre-attentive processing in visual search tasks can be directly linked to texture boundary processing as revealed by recent EEG studies on visual search.     

Pre-attentive segmentation and correspondence in stereo

Traditional stereo grouping models have focused on the problem of stereo correspondence between monocular inputs. Recent physiological data revealed that the disparity selective V2 cells increase their responses when (random-dot stereograms) stimuli within their receptive fields are at or near the boundary of a depth surface. Such highlights to depth (non-luminance) edges are seemingly not computationally required for the correspondence problem. Computationally, these highlights make the boundaries of a depth surface more salient, serving pre-attentive segmentation (between depth planes) and attracting visual attention. In special cases, they enable the psychophysically observed perceptual pop-out of a target from a background of visually identical distractors at a different depth. To achieve the highlights, mutual inhibition between disparity selective cells that are tuned to the same or similar depths is required. However, such mutual inhibition would impede the computation for the correspondence problem, which requires mutual excitation between the same cells. In this work, I introduce a computational model that, I believe, is the first to address both stereo correspondence and pre-attentive stereo segmentation. The computational mechanisms in the model are based on intracortical interactions in V2. I will demonstrate that the model captures the following physiological and psychophysical phenomena: (i) depth-edge highlighting; (ii) disparity capture; (iii) pop-out; and (iv) transparency.     

Mechanisms of neuronal computation in Mammalian visual cortex

Orientation selectivity in the primary visual cortex (V1) is a receptive field property that is at once simple enough to make it amenable to experimental and theoretical approaches and yet complex enough to represent a significant transformation in the representation of the visual image. As a result, V1 has become an area of choice for studying cortical computation and its underlying mechanisms. Here we consider the receptive field properties of the simple cells in cat V1-the cells that receive direct input from thalamic relay cells-and explore how these properties, many of which are highly nonlinear, arise. We have found that many receptive field properties of V1 simple cells fall directly out of Hubel and Wiesel's feedforward model when the model incorporates realistic neuronal and synaptic mechanisms, including threshold, synaptic depression, response variability, and the membrane time constant.     

Motion edges and regions guide image segmentation by colour.

Image segmentation is an important early stage in visual processing in which the visual system groups together parts of the image that belong together, prior to or in conjunction with object recognition. Two principal processes may be involved in image segmentation: an edge-based process that uses feature contrasts to mark boundaries of coherent regions, and a region-based process that groups similar features over a larger scale. Earlier, we have shown that motion and colour interact strongly in image segmentation by the human visual system. Here we explore the nature of this interaction in terms of edge- and region-based processes. We measure performance on a region-based colour segmentation task in the presence of distinct types of motion information, in the form of edges and regions which in themselves do not reveal the location of the colour target. The results show that both motion edges and regions may guide the integrative process required for this colour segmentation task. Motion edges appear to act by delimiting areas over which to integrate colour information, whereas motion similarities define primitive surfaces within which colour grouping and segmentation processes are deployed.     

Dimensional reduction of a V1 ring model with simple and complex cells

In this paper, we extend a framework for constructing low-dimensional dynamical systems models of mammalian primary visual cortex to a cortical network model that incorporates the full nonlinear effects of complex cells. The procedure consists of capturing the essential dynamics in a low-dimensional subspace using empirical methods, then recasting the equations in the reduced vector space. Previously, we considered visual cortical network models consisting of only simple cells with nearly linear responses to external stimuli. Here we show that fully nonlinear effects can be incorporated by examining the dimensional reduction of an idealized ring model of V1 with both simple and complex cells. We found it expedient to divide the subspace into four separate neuronal populations: excitatory simple, excitatory complex, inhibitory simple and inhibitory complex. In order to reproduce the fluctuation-driven dynamics in this reduced space, we incorporated (1) white noises with different intensities into individual neuronal populations, and (2) firing rate estimates to capture the probability of firing due to subthreshold fluctuations. With a more accurate, fitted connectivity, our modified dimensional reduced models can reproduce the firing rates, circular variances and modulation ratios observed in the original ring model.     

Extended difference-of-Gaussians model incorporating cortical feedback for relay cells in the lateral geniculate nucleus of cat

A striking feature of the organization of the early visual pathway is the significant feedback from primary visual cortex to cells in the dorsal lateral geniculate nucleus (LGN). Despite numerous experimental and modeling studies, the functional role for this feedback remains elusive. We present a new firing-rate-based model for LGN relay cells in cat, explicitly accounting for thalamocortical loop effects. The established DOG model, here assumed to account for the spatial aspects of the feedforward processing of visual stimuli, is extended to incorporate the influence of thalamocortical loops including a full set of orientation-selective cortical cell populations. Assuming a phase-reversed push-pull arrangement of ON and OFF cortical feedback as seen experimentally, this extended DOG (eDOG) model exhibits linear firing properties despite non-linear firing characteristics of the corticothalamic cells. The spatiotemporal receptive field of the eDOG model has a simple algebraic structure in Fourier space, while the real-space receptive field, as well as responses to visual stimuli, are found by evaluation of an integral. As an example application we use the eDOG model to study effects of cortical feedback on responses to flashing circular spots and patch-grating stimuli and find that the eDOG model can qualitatively account for experimental findings.     

A regression model for carpal tunnel syndrome

The purpose of this study was to determine whether a logistic regression model for the diagnosis of carpal tunnel syndrome (CTS) could be developed. Forty-eight variables were initially identified, for the 28 CTS and 34 non-CTS subjects, including 28 measures of nerve function, 6 anatomical measurements, 8 variables relating to disease symptoms, and 6 variables relating to physical attributes. An a priori clustering procedure was used to establish groups for the principal components analyses. The first principal component of each cluster was then used in a backward, stepwise logistic regression analysis. The best combination of candidate variables, as identified by the regression equation, was Raynaud's symptoms and median nerve motor function. The results of this study indicate that a model for CTS can be generated from a set of variables and that a linear combination of variables representing nerve function is closely associated with conduction decrements resulting from CTS.     

Computational models of visual neurons specialised in the detection of periodic and aperiodic oriented visual stimuli: bar and grating cells

Computational models of periodic- and aperiodic-pattern selective cells, also called grating and bar cells, respectively, are proposed. Grating cells are found in areas V1 and V2 of the visual cortex of monkeys and respond strongly to bar gratings of a given orientation and periodicity but very weakly or not at all to single bars. This non-linear behaviour, which is quite different from the spatial frequency filtering behaviour exhibited by the other types of orientation-selective neurons such as the simple cells, is incorporated in the proposed computational model by using an AND-type non-linearity to combine the responses of simple cells with symmetric receptive field profiles and opposite polarities. The functional behaviour of bar cells, which are found in the same areas of the visual cortex as grating cells, is less well explored and documented in the literature. In general, these cells respond to single bars and their responses decrease when further bars are added to form a periodic pattern. These properties of bar cells are implemented in a computational model in which the responses of bar cells are computed as thresholded differences of the responses of corresponding complex (or simple) cells and grating cells. Bar and grating cells seem to play complementary roles in resolving the ambiguity with which the responses of simple and complex cells represent oriented visual stimuli, in that bar cells are selective only for form information as present in contours and grating cells only respond to oriented texture information. The proposed model is capable of explaining the results of neurophysiological experiments as well as the psychophysical observation that the perception of texture and the perception of form are complementary processes. Received: 4 June 1996 / Accepted in revised form: 7 October 1996