Spike Encoding with Optic Sensory Neurons Enable a Pulse Coupled Neural Network for Ultraviolet Image Segmentation.

Neuromorphic computing [1], inspired by the energy-efficient and parallelized architecture of the human brain, represents a paradigm shift in computing. This approach not only drives innovative hardware designs but also inspires novel network architectures like spiking neural networks (SNNs), which process information as discrete spike events. However, scalability in neuromorphic systems is constrained by the large hardware footprint. Realizing multimodal functionality using the same set of hardware primitives can offer a promising solution to these limitations.In this work, we present a dual encoder-neuron circuit based on a nanoporous graphene (NPG) memristor device [2]. The graphene leaky integrate-and-fire (GLIF) circuit has a versatile multimodal design that combines the roles of a LIF neuron and a spike encoder within the same architecture. This design mimics the dual functionality of biological neurons, which encode and process signals locally. The spike encoding capability of the circuit is further revealed using single and double layer SNN networks for pattern recognition tasks using the Modified National Institute of Standards and Technology (MNIST) dataset.Inset of Figure 1(a) shows the schematic of the fabricated NPG device with a channel length of 100 µm. The device has a lateral structure with NPG channel and gold (Au) electrodes. Figure 1(a) shows the current-voltage (I-V) characteristics of the fabricated NPG device which shows a threshold switching behavior with a wide hysteresis window and a threshold voltage (Vth) of 4.9 V. A behavioral SPICE model is developed based on a voltage-controlled switch model. It is seen from Figure 1(a) that the I-V curves simulated using the SPICE model matches the experimentally observed device switching characteristics. This reveals that the simplified SPICE model can be a good substitute for complex Verilog based models. Based on the SPICE model, we designed a simple neuron circuit with a resistor (Rs, 12 kΩ) in series with a parallel combination of a capacitor (Cm, 10 nF) and the NPG device (SPICE model). Figure 1(b) shows the variation of the membrane potential (Vm) and the output spike current in response to the input pulse voltage (Vin, 6 V). Vm shows a clear leaky integration behavior and full reset to zero voltage after the spike response. These are essential attributes of a bioplausible LIF neuron circuit. In order to develop the multimodal GLIF circuit, the output spike frequency of the LIF neuron circuit is plotted as a function of the Rs values. This mapping allows pixel values from MNIST input images to be directly translated into normalized conductance values (G/G0). Hence, each pixel is represented as a normalized conductance value, establishing a clear correlation between pixel intensity and spike frequency. This core idea is further used to develop the GLIF encoder based on the same neuron circuit as before. Based on the dual encoder-neuron circuit, we designed a single layer SNN for recognising the handwritten digits from the MNIST dataset. Figure 1(c) shows the spike-encoded images for the MNIST digits “2” and “5”. The probability of spiking in each timestep is proportional to the spiking frequency. The spike encoded images at each time step reveal the contours of the original input images using sparse representation. These sparse matrices highlight the energy efficiency and computational advantages of the GLIF system. The performance of the dual encoder-neuron system is evaluated by training the single layer SNN using both the GLIF encoder and neuron circuit as integral components. The single layer SNN recorded a high pattern recognition accuracy of 90.77% which is comparable to the purely software-based implementation (92.69%). To explore the scalability of the GLIF system, we implemented a double-layer SNN using the GLIF system, achieving a high recognition accuracy of 97.37%. The GLIF circuit exemplifies how reconfigurable primitive components can drive the development of scalable neuromorphic hardware.

Image segmentation is a very important aspect in the field of image processing. According to certain principles, it will be divided into a number of small regional disjoint. Small region is a sense sets of common attributes of the connectivity pixels. The so-called connectivity which refers to the fact that any two points in the pool existing a connectivity path between all belong to the set. Pulse coupled neural networks was proposed based on the brain simultaneously pulsed release phenomenon of the visual cortex of the cats, monkeys, and other animals, the similar input data can be outbreak synchronized through interrelated neurons. The digital image can be processed by pulse coupled neural networks, the pixels of space based approach and brightness can be made up to a group. Pulse coupled neural networks has a good nature in image processing applications, which can effectively for the various aspects of image processing, and it can be used in the image processing system. Experimental results show that the method can achieve better image segmentation and have a common applicability.

In this paper, a review of parameter optimization methods of pulse-coupled neural networks (PCNNs) is presented. Considering that PCNN has been used in image processing for many years, the aim of this paper was to provide an overview of the work that has been done and to serve as a useful reference for those who are looking for PCNN parameter optimization methods and those who are researching PCNN applications for a specific field. This paper first briefly reviews the PCNN model, including the standard PCNN and several variants of PCNN. Then, we emphasize the optimization methods for PCNN’s parameters, describing three types of parameter optimization methods in detail. Next, the paper summarizes the applications of the optimized models of PCNN with adaptive parameters in image segmentation, image fusion, image denoising and edge detection.

Image analysis for airborne reconnaissance and missile applications

It needs set parameters on image segmentation based on PCNN (Pulse Coupled Neural Network) now. This paper points out the new method for medical image segmentation based on improved PCNN and Tsallis entropy. The new methods can automatically segment the medical images without selecting the PCNN parameters. It gets the best results with combining with the Tsallis entropy. The new method is very useful for PCNN application in the medical images segmentation.

Visual object tracking is a fundamental research topic in computer vision. In this paper, we proposed a novel hybrid tracking method based on Pulse Coupled Neural Network (PCNN) and Multiple Instance Learning (MIL). Most modern trackers may be inaccurate when the training samples are imprecise which causes drift. To resolve these problems, MIL method is introduced into the tracking task, which can alleviate drift to some extent. However, the MIL tracker may detect the positive sample that is less important. PCNN is different from traditional artificial neural networks, which can be applied in many image processing fields, such as image segmentation. So, the PCNN was employed as sample detector which can know the most important sample when training the classifier. Then, a more robust and much faster tracker is proposed to approximately maximize the bag likelihood function. Empirical results on a large set of sequences demonstrate the superior performance of the proposed approach in robustness, stability and efficiency to state-of-the-art methods in the literature.

This paper discusses some of the advantages and disadvantages of pulse-coupled neural networks (PCNNs) for performing image segmentation in the realm of medical diagnostics. PCNNs were tested with magnetic resonance imagery (MRI) of the brain and abdominal region and nuclear scintigraphic ventilation/perfusion imagery of the lungs (V/Q scans). Our preliminary results show that PCNNs do well at contrast enhancement. They also do well at image segmentation when each segment is approximately uniform in intensity. However, there are limits to what PCNNs can do. For example, when intensity significantly varies across a single segment, that segment does not properly separate from other objects. Another difficulty is finding the optimum parameter values so that a uniform response is achieved over a set of imagery. Also, a set of parameters that properly segments objects in one image is sometimes unsuccessful on a similar image.

Pulse-coupled neural network (PCNN), which simulates the synchronous oscillation phenomenon in the visual cortex of small mammals, has become a useful model for image processing. In the model, several parameters were usually required to properly set for adjusting the behavior of neurons. However, undesired behavior may occur owing to inappropriate parameters setting. To alleviate this problem, we propose to simplify some parameters of PCNN, and apply it into image segmentation. First, exponential delay factors are abandoned for adjusting the neuron input, and the neural input is then associated with image information as well as pulse output. In addition, neural threshold inherent in PCNN is simplified as an adaptive threshold related to image properties, allowing our model to easily alter the behavior of neurons. Particularly, the characteristic of synchronous pulse is thereby kept by introducing a fuzzy clustering method, instead of linking coefficient for grouping pixels with similarity and spatial proximity through iterative computation. Experimental results on synthetic and real infrared images show that the proposed model has high performance of segmentation. Furthermore, our model has better adaptability for segmenting real-world images when compared with several existing PCNN-based methods and some classic segmentation methods.

Pulse-Coupled Neural Network (PCNN) is a new kind of network, which is constructed by simulating the activities of the mammal's visual cortex neurons. PCNN is suitable for image processing, including segmentation, image fusion, edge extraction. In this paper, a new method is proposed which is the use of Bidirectional search Pulse-Coupled Neural Network (BPCNN) in the color image segmentation. The validity of this method is well verified by the analysis and experiments.

It is the key step of automatic apple picking and sorting to segment of the apple image in natural environment using image information processing technology. In order to improve the intellectualization of apple picking process and effectively extract the target of image segmentation, a novel image segmentation method of Huaniu apple based on pulse coupled neural network (PCNN) and watershed algorithm is proposed. Firstly, the PCNN model is improved and optimized on the basis of image RGB to L*A*B space conversion. Then, the improved PCNN combined with watershed algorithm was used to segment the apple image on considering the image texture and color feature information. The theoretical analysis and experimental results show that the novel method is effectively able to segment target image, and can preserve the image details and the edges as well. The proposed algorithm has better subjective vision effect and objective quality than the traditional watershed and other related algorithms.

Light field can generate a set of refocused images by integrating rays into any image plane. It is useful to segment focus region from the refocused image because the focus region contains depth clues. However, to the best of our knowledge, focus-region segmentation for light field images has not been reported. We firstly transform the refocused image into the space of neuron ignition sequence with pulse coupled neural networks (PCNN). Then the model of neuron ignition sequence and the classification criteria are established to divide pixels into different regions. Finally, the clearest region is selected as the segmentation result by the sliding window technology. In this paper, we analyze the features of focus/defocus information in the space of neuron ignition sequence, and propose some simple criteria for pixel classification. The proposed algorithm and other two typical segmentation algorithms for low depth-of-field images are tested on the data set of light field images. The experimental results show that the three algorithms have similar error rates of focused-region segmentation, but the computational time of ours is one order of magnitude shorter than that of the other two.

Neuromorphic computing provides a promising alternative to conventional von Neumann architectures by addressing memory bottlenecks and energy inefficiencies. In this work, a hybrid memristor‐based neuromorphic system is presented, comprising a volatile Pt/V/AlO x /Pt threshold switching memristor (TSM) as the spiking neuron and a non‐volatile Ta/ZrO x /Pt resistive switching memristor (RSM) as the artificial synapse. The neuron circuit realizes a leaky‐integrate‐and‐fire (LIF) model with tunable synaptic weights and temporal encoding functions. The artificial synapses exhibit multi‐level resistance states and retention characteristics, ensuring long‐term conductance retention. By integrating the neuron circuit with multiple artificial synapses, it is demonstrated that membrane potential (V M ) dynamics are jointly governed by input strength and timing, thereby enabling precise modulation of spike generation. In addition, the TSM‐based neuron can also serve as a spike encoder, where the output firing rate or time‐to‐first‐spike (TTFS) is directly modulated by the amplitude of the input signal. This dual functionality allows the same circuit to function as both a computational unit and an analog‐to‐spike encoder. Overall, the circuit replicates essential neuronal dynamics, including subthreshold integration, temporal summation, and synchronized firing, validating its potential for scalable neuromorphic hardware implementation.

Mega voltage X-ray images are often characterized by low contrast, low resolution and image blurring. We explore efficient image segmentation of different types of mega voltage X-ray images using a specific neural processing technique called pulse coupled neural network (PCNN). We also study the usefulness of PCNN in ambient noise removal for the high voltage X-ray images.

MH-Net: Model-data-driven hybrid-fusion network for medical image segmentation

Despite the maturity and availability of speech recognition systems, there are few available spiking speech recognition tasks that can be implemented with current neuromorphic systems. The methods used previously to generate spiking speech data are not capable of encoding speech in real-time or encoding very large modern speech datasets efficiently for input to neuromorphic processors. The ability to efficiently encode audio data to spikes will enable a wider variety of spiking audio datasets to be available and can also enable algorithmic development of real-time neuromorphic automatic speech recognition systems. Therefore, we developed speech2spikes, a simple and efficient audio processing pipeline that encodes recorded audio into spikes and is suitable for real-time operation with low-power neuromorphic processors. To demonstrate the efficacy of our method for audio to spike encoding we show that a small feed-forward spiking neural network trained on data generated with the pipeline achieves accuracy on the Google Speech Commands recognition task, exceeding the state-of-the art set by Spiking Speech Commands, a prior spiking encoding of the Google Speech Commands dataset, by over 10%. We also demonstrate a proof-of-concept real-time neuromorphic automatic speech recognition system using audio encoded with speech2spikes streamed to an Intel Loihi neuromorphic research processor.