Biological Retina Research Articles

Cybernetics of Vision Systems: Toward an Understanding of Putative Functions of the Outer Retina

The retina still poses many structural and computational questions. Structurally, for example, it is not yet clear how many distinct horizontal cell (HC) types the primate retina contains and what the exact patterns of connections between photoreceptors (PRs) and HCs consist of. Computationally, it is not yet clear, for instance, what functions are present and how they are being implemented. This paper proposes a model (a linear recurrent neural network defined by 31 parameters) of the outer retina and an optimization methodology that hopes to shed some light on these questions. This paper shows that a simplified model of the outer retina can implement several low-level visual functions involving the modulation of noise, brightness, contrast, saturation, and even color. The results demonstrate that contrast control functions can be implemented with a minimum of two HC types and that spectral specificity between PRs and HCs is a common and important feature. It is also shown that several different spectrally specific patterns can emerge in order to implement the same function. One interesting microcircuit that naturally emerged from our experiments involves nonblurry denoising via interchromatic gap junctions and compensatory resaturation via HC circuits, a strategy that we hypothesize to exist in some biological retinae.

Visual simulation of retinal images through microstructures

The space distribution of photoreceptors is a main topic in artificial retina design procedures. A fundamental question is how different retina architectures reproduce the same image and which their properties are? This paper presents a new method to emulate the visual response of biological or artificial retinas. The proposed method can predict the vision output image of any retina with known photoreceptors distribution. A simple method is proposed that considers only two parameters to simulate the output image of any retina architecture: the distribution of the photoreceptors in the retina and the estimated light response of its photoreceptors. The distribution of the photoreceptors along the retina is modeled by geometrical figures that let define a set of many different outer trajectories. The paper discusses also a new method to organize the layout of any retinal architecture design using a set of basic square vertical phototransistors cells according with a previously defined photoreceptors distribution. The Lena image is used to predict the degree of resolution at the central area and how it is faded towards the periphery.

Penrose Pixels for Super-Resolution

We present a novel approach to reconstruction-based super-resolution that uses aperiodic pixel tilings, such as a Penrose tiling or a biological retina, for improved performance. To this aim, we develop a new variant of the well-known error back projection super-resolution algorithm that makes use of the exact detector model in its back projection operator for better accuracy. Pixels in our model can vary in shape and size, and there may be gaps between adjacent pixels. The algorithm applies equally well to periodic or aperiodic pixel tilings. We present analysis and extensive tests using synthetic and real images to show that our approach using aperiodic layouts substantially outperforms existing reconstruction-based algorithms for regular pixel arrays. We close with a discussion of the feasibility of manufacturing CMOS or CCD chips with pixels arranged in Penrose tilings.

FPGA-Based Real Time, Multichannel Emulated-Digital Retina Model Implementation

The function of the low-level image processing that takes place in the biological retina is to compress only the relevant visual information to a manageable size. The behavior of the layers and different channels of the neuromorphic retina has been successfully modeled by cellular neural/nonlinear networks (CNNs). In this paper, we present an extended, application-specific emulated-digital CNN-universal machine (UM) architecture to compute the complex dynamic of this mammalian retina in video real time. The proposed emulated-digital implementation of multichannel retina model is compared to the previously developed models from three key aspects, which are processing speed, number of physical cells, and accuracy. Our primary aim was to build up a simple, real-time test environment with camera input and display output in order to mimic the behavior of retina model implementation on emulated digital CNN by using low-cost, moderate-sized field-programmable gate array (FPGA) architectures.

Retina-Motivated CMOS Vision Chip Based on Column Parallel Architecture and Switch-Selective Resistive Network

m double-poly four-metal complementary metal-oxide-semiconductor technology. It mimics the edge detection mechanism of a biological retina. This type of vision chip offer several advantages including compact size, high speed, and dense system integration. Low resolution and relatively high power consumption are common limitations of these chips because of their complex circuit structure. We have tried to overcome these problems by rearranging and simplifying their circuits. A vision chip of 160

Virtual Retina: A biological retina model and simulator, with contrast gain control

We propose a new retina simulation software, called Virtual Retina, which transforms a video into spike trains. Our goal is twofold: Allow large scale simulations (up to 100,000 neurons) in reasonable processing times and keep a strong biological plausibility, taking into account implementation constraints. The underlying model includes a linear model of filtering in the Outer Plexiform Layer, a shunting feedback at the level of bipolar cells accounting for rapid contrast gain control, and a spike generation process modeling ganglion cells. We prove the pertinence of our software by reproducing several experimental measurements from single ganglion cells such as cat X and Y cells. This software will be an evolutionary tool for neuroscientists that need realistic large-scale input spike trains in subsequent treatments, and for educational purposes.

Retina simulation using cellular automata and GPU programming

This article shows how the architectural modelization of biological retina allows real-time performances, on standard widespread computing systems. First, we describe the biological retina with regard to its pipeline architecture, detailing its layer behaviours and properties. Then we propose a corresponding pipelined model of artificial retina based on cellular automata. In this work, the main innovation is the computing method based on the programming of a personal computer graphical card using OpenGL shading language. The last section demonstrates the efficiency of our model through numerical and graphical results. We lay particular emphasis on the fact that our direct implementation of the Graphical Processor Unit (GPU) provides computation power about 20 times as fast as conventional programming.

A novel adaptive logarithmic digital pixel sensor

A major problem associated with complementary metal-oxide-semiconductor and charge couple device imagers is their limited dynamic range (DR), typically 60-70 dB. This falls far short of covering the wide illumination ranges found in natural scenes (typically 120-140 dB). Biological retinas are known to feature adaptive, logarithmic-type responses enabling them to cover a very wide DR, without compromising the resolution. This letter presents a novel way to realize such an adaptive logarithmic response by combining a digital time domain vision sensor and a simple adaptive digital quantizer. This letter presents the theory and experimental results for an adaptive logarithmic response sensor featuring over 100-dB DR

A Bio-Inspired 128 × 128 Complementary Metal–Oxide–Semiconductor Vision Chip for Edge Detection with Signal Processing Circuit Separated from Photo-Sensing Circuit

In this paper, a vision chip for edge detection based on the structure of biological retina is introduced. The key advantage of retinal structure is speed of operation. However, bio-inspired vision chips have suffered from low resolution which is caused by their complex circuit structure. To sufficiently improve the resolution for real application, the circuits for analog processing were separated from circuits for image capturing. In addition, we compensated the low speed problem of operation which is caused by bottleneck of data transfer between photo-sensors and analog processing circuit by adding a reset decoder. The vision chip was fabricated using 0.35 μm 2-poly 4-metal complementary metal---oxide---semiconductor technology. Using the vision chip, we could obtain a contrast-enhanced image without any other cost-increase for fabrication of chips. Then, the edge image was easily achieved by thresholding the previous contrast-enhanced image.

수광 회로와 윤곽 검출 회로의 분리를 통한 윤곽 검출용 시각칩의 해상도 향상

Resolution of an image sensor is very significant parameter to improve. It is hard to improve the resolution of the CMOS vision chip for edge detection based on a biological retina using a resistive network because the vision chip contains additional circuits such as a resistive network and some processing circuits comparing with general image sensors such as CMOS image sensor (CIS). In this paper, we proved the problem of low resolution by separating photo-sensing and signal processing circuits. This type of vision chips occurs a problem of low operation speed because the signal processing circuits should be commonly used in a row of the photo-sensors. The low speed problem of operation was proved by using a reset decoder. A vision chip for edge detection with <TEX>$128{\times}128$</TEX> pixel array has been designed and fabricated by using <TEX>$0.35{\mu}m$</TEX> 2-poly 4-metal CMOS technology. The fabricated chip was integrated with optical lens as a camera system and investigated with real image. By using this chip, we could achieved sufficient edge images for real application.

픽셀의 고정 패턴 잡음을 감소시킨 윤곽 검출용 시각칩

When fabricating a vision chip, we should consider the noise problem, such as the fixed pattern noise(FPN) due to the process variation. In this paper, we propose an edge-detection circuit based on biological retina using the offset-free column readout circuit to reduce the FPN occurring in the photo-detector. The offset-free column readout circuit consists of one source follower, one capacitor and five transmission gates. As a result, it is simpler and smaller than a general correlated double sampling(CDS) circuit. A vision chip for edge detection has been designed and fabricated using <TEX>$0.35\;{\mu}m$</TEX> 2-poly 4-metal CMOS technology, and its output characteristics have been investigated.

Implantable neuromorphic vision chips

The implantable neuromorphic vision chips are analysed and implemented. These proposed analogue integrated circuits are to implement four cell functions of the retina for implant. These retinal cells are photoreceptor cells, horizontal cells, bipolar cells, and ganglion cells. All simulation results are successfully verified and consistent with researches on biological retinas.

CMOS realization of a 2-layer CNN universal machine chip.

Some features of the biological retina can be modelled by a 2-layer cellular neural network (CNN) composed of locally connected elementary nonlinear processors. In order to explore these complex spatiotemporal dynamics for image processing, a prototype chip has been designed and fabricated in a 0.5 microm CMOS technology. Design challenges, trade-offs, the building blocks and the tests results for this system with 0.5 x 10(6) transistors, most of them operating in analog mode, are presented in this paper.

Optical output silicon retina chip

We describe a unique implementation of an optical-in/optical- out focal plane processor whose functionality is derived from a biological retina. The retina chip has increased control over the output through a novel differential difference amplifier and a clocked analog scheme. Pro- cessing is done on-chip to produce an edge-detected version of the op- tical input signal. However, the output signal is also optical, enabled by a liquid crystal spatial light modulator on the chip. We present experimen- tal results from a 19319 array and demonstrate how the edge informa- tion can be controlled on-chip and read out in parallel from each process- ing cell. © 2003 Society of Photo-Optical Instrumentation Engineers.

The retinomorphic approach: pixel-parallel adaptive amplification, filtering, and quantization

I describe a vision system that uses neurobiological principles to perform all four major operations found in biological retinae: (1) continuous sensing for detection, (2) local automatic gain control for amplification, (3) spatiotemporal bandpass filtering for preprocessing, and (4) adaptive sampling for quantization. All four operations are performed at the pixel level. The system includes a random-access time-division multiplexed communication channel that reads out asynchronous pulse trains from a 64×64 pixel array in the imager chip, and transmits them to corresponding locations on a second chip that has a 64×64 array of integrators. Both chips are fully functional. I compare and contrast the design principles of the retina with the standard practice in imager design and analyze the circuits used to amplify, filter, and quantize the visual signal, with emphasis on the performance trade-offs inherent in the circuit topologies used.

Visual information processing in primate cone pathways. II. Experiments

For pt.I. see ibid., p.259-74. This paper presents the results of experiments performed to validate the computer retina model presented in part I. Experiments commonly performed by electrophysiologists on biological retinas are simulated and the computer retina outputs compared with published recordings of cells in biological (principally primate) retinas. Experiments with more complex stimuli further reveal how the computer retina enhances spatio-temporal contrast information and adapts to a wide range of illumination levels much like the primate retina.

Visual information processing in primate cone pathways. I. A model

At the retinal level, the strategies utilized by biological visual systems allow them to outperform machine vision systems, serving to motivate the design of electronic or "smart" sensors based on similar principles. Design of such sensors in silicon first requires a model of retinal information processing which captures the essential features exhibited by biological retinas. In this paper, a simple retinal model is presented, which qualitatively accounts for the achromatic information processing in the primate cone system. The computer retina model exhibits many of the properties found in biological retinas such as data reduction through nonuniform sampling, adaptation to a large dynamic range of illumination levels, variation of visual acuity with illumination level, and enhancement of spatiotemporal contrast information. The main emphasis of the model presented here is to demonstrate how different adaptation mechanisms play a role in extending the operating range of the primate retina.

A retinomorphic vision system

This system uses neurobiological principles to accomplish the major operations of biological retinae: continuous sensing, local automatic gain control, spatiotemporal bandpass filtering, and adaptive sampling.

An artificial retina chip made of a 128 x 128 pn-np variable-sensitivity photodetector array

An artificial retina chip is an image detection device that, like biological retinas, can extract essential features directly from projected images. This paper presents an artificial retina chip consisting of a 128/spl times/128 array of pn-np variable-sensitivity photodetectors (VSPD's). This photodetector (PD) utilizes pn junctions and thus avoids the rather inflexible memory effect encountered in our previously reported device, which used Schottky contacts in a metal-semiconductor-metal (MSM) configuration. We introduce a novel "filter division" method for arbitrary two-dimensional (2-D) filter operation, and successfully employ it in a 2-D edge extraction task.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>