A Mathematical Model of Retinal Ganglion Cells and Its Applications in Image Representation

Binocular Matching Model Based on Hierarchical V1 and V2 Receptive Fields With Color, Orientation, and Region Feature Information

Binocular matching models serve as the core component in most stereo visual aid systems developed for people with visual impairments. However, purely computational models lack a neuro-biological basis for explaining the phenomena observed in neuro-biology, and therefore offer no support for the development of bioengineering applications, and are overly complex for hardware implementation. In contrast, existing neurobiological models suffer from low matching calculation accuracy. Therefore, the present work proposes a novel binocular matching model based on the receptive field of simple cells rather than on image pixels, and thereby incorporates neurobiological structure, reduces hardware complexity, has enough accuracy and can be used in visual aid system. The proposed model is employed to calculate and optimize the binocular disparity via a cost function. Specifically, we simulate the functions and structures of V1 and V2 neurons according to the discoveries of modern neurobiology. Accordingly, the receptive fields of V1 layer neurons are aggregated to obtain the receptive fields of the V2 layer, and the disparity is obtained in the V2 layer. The accuracy of the proposed model is verified by comparisons of the disparity results obtained using the proposed model with those obtained using other neurobiological model, and thereby demonstrates that the model can guide the design of visual aid systems.

A Signal-Processing Neural Model Based on Biological Retina

Image signal processing has considerable value in artificial intelligence. However, due to the diverse disturbance (e.g., color, noise), the image signal processing, especially the representation of the signal, remains a big challenge. In the human visual system, it has been justified that simple cells in the primary visual cortex are obviously sensitive to vision signals with partial orientation features. In other words, the image signals are extracted and described along the pathway of visual processing. Inspired by this neural mechanism of the primary visual cortex, it is possible to build an image signal-processing model as the neural architecture. In this paper, we presented a method to process the image signal involving a multitude of disturbance. For image signals, we first extracted 4 rivalry pathways via the projection of color. Secondly, we designed an algorithm in which the computing process of the stimulus with partial orientation features can be altered into a process of analytical geometry, resulting in that the signals with orientation features can be extracted and characterized. Finally, through the integration of characterizations from the 4 different rivalry pathways, the image signals can be effectively interpreted and reconstructed. Instead of data-driven methods, the presented approach requires no prior training. With the use of geometric inferences, the method tends to be interpreted and applied in the signal processor. The extraction and integration of rivalry pathways of different colors allow the method to be effective and robust to the signals with the image noise and disturbance of colors. Experimental results showed that the approach can extract and describing the image signal with diverse disturbance. Based on the characterization of the image signal, it is possible to reconstruct signal features which can effectively represent the important information from the original image signal.

Approaching Retinal Ganglion Cell Modeling and FPGA Implementation for Robotics

Taking inspiration from biology to solve engineering problems using the organizing principles of biological neural computation is the aim of the field of neuromorphic engineering. This field has demonstrated success in sensor based applications (vision and audition) as well as in cognition and actuators. This paper is focused on mimicking the approaching detection functionality of the retina that is computed by one type of Retinal Ganglion Cell (RGC) and its application to robotics. These RGCs transmit action potentials when an expanding object is detected. In this work we compare the software and hardware logic FPGA implementations of this approaching function and the hardware latency when applied to robots, as an attention/reaction mechanism. The visual input for these cells comes from an asynchronous event-driven Dynamic Vision Sensor, which leads to an end-to-end event based processing system. The software model has been developed in Java, and computed with an average processing time per event of 370 ns on a NUC embedded computer. The output firing rate for an approaching object depends on the cell parameters that represent the needed number of input events to reach the firing threshold. For the hardware implementation, on a Spartan 6 FPGA, the processing time is reduced to 160 ns/event with the clock running at 50 MHz. The entropy has been calculated to demonstrate that the system is not totally deterministic in response to approaching objects because of several bioinspired characteristics. It has been measured that a Summit XL mobile robot can react to an approaching object in 90 ms, which can be used as an attentional mechanism. This is faster than similar event-based approaches in robotics and equivalent to human reaction latencies to visual stimulus.

A Retinotopic Spiking Neural Network System for Accurate Recognition of Moving Objects Using NeuCube and Dynamic Vision Sensors

This paper introduces a new system for dynamic visual recognition that combines bio-inspired hardware with a brain-like spiking neural network. The system is designed to take data from a dynamic vision sensor (DVS) that simulates the functioning of the human retina by producing an address event output (spike trains) based on the movement of objects. The system then convolutes the spike trains and feeds them into a brain-like spiking neural network, called NeuCube, which is organized in a three-dimensional manner, representing the organization of the primary visual cortex. Spatio-temporal patterns of the data are learned during a deep unsupervised learning stage, using spike-timing-dependent plasticity. In a second stage, supervised learning is performed to train the network for classification tasks. The convolution algorithm and the mapping into the network mimic the function of retinal ganglion cells and the retinotopic organization of the visual cortex. The NeuCube architecture can be used to visualize the deep connectivity inside the network before, during, and after training and thereby allows for a better understanding of the learning processes. The method was tested on the benchmark MNIST-DVS dataset and achieved a classification accuracy of 92.90%. The paper discusses advantages and limitations of the new method and concludes that it is worth exploring further on different datasets, aiming for advances in dynamic computer vision and multimodal systems that integrate visual, aural, tactile, and other kinds of information in a biologically plausible way.

A Genetic-Algorithm-Based Explicit Description of Object Contour and its Ability to Facilitate Recognition

Shape representation is an extremely important and longstanding problem in the field of pattern recognition. Closed contour, which refers to shape contour, plays a crucial role in the comparison of shapes. Because shape contour is the most stable, distinguishable, and invariable feature of an object, it is useful to incorporate it into the recognition process. This paper proposes a method based on genetic algorithms. The proposed method can be used to identify the most common contour fragments, which can be used to represent the contours of a shape category. The common fragments clarify the particular logics included in the contours. This paper shows that the explicit representation of the shape contour contributes significantly to shape representation and object recognition.

A Line-Context Based Object Recognition Method

The shape or contour of an object is usually stable and persistent, so it is a good basis for invariant recognition. For this purpose, two problems must be handled. The first is obtaining clean edges and the other is organizing those edges into a structured form so that they can be manipulated easily. We apply a bio-inspired orientation detection algorithm because it can output a fairly clean set of lines, and all lines are in the form of vectors instead of pixels. This line representation is efficient. We decompose them into several slope-depended layers and then create a hierarchical partition tree to record their geometric distribution. Based on the similarity of trees, a rough classification of objects can be realized. However, for an accuracy recognition, we design a moment-based measure to describe the detail layout of lines in a layer and then re-describe image by Hu's moment invariants. The experimental results suggest that the representation efficiency enabled by simple cell's neural mechanism and application of multi-layered representation schema can simplify the complexity of the algorithm. This proves that line-context representation greatly eases subsequent shape-oriented recognition.