Memristive continuous Hopfield neural network circuit for image restoration

Based on continuous Hopfield neural network (CHNN), a new alternative is developed for solving linear quadratic (LQ) optimal control problem of discrete-time systems. In this method, the LQ performance index is transformed into the energy function of CHNN, and the control sequence into the output vector of the neurons of CHNN. As a result, solving LQ dynamic optimization problem is equivalent to operating associated CHNN from its initial state to the terminal state. The stable output vector of CHNN represents the optimal control sequence. Because CHNN works in parallel and is of real-time characteristic, the present method is easier to satisfy the requirement of real-time control and will be promising in application.

The paper proposes a new method to blindly detect QPSK signals of MIMO systems by using continuous Hopfield neural network (HNN). The new method successfully constructs a weighted matrix of HNN, which reduces the number of equilibrium points of HNN significantly, so that the speed to blindly detect signals is increased as well as the bit error rate (BER) is lowered. The new method detects multi-user signals with serial processing, in which the method utilizes those user signal vectors having been detected to extend the complementary subspace consisting of received data vectors in order to recover other user signals to be detected. The performance of the proposed HNN-based method is evaluated via computer simulations and compared with the existing approaches. Results are shown that the new method has better performances than others.

Due to their synaptic-like characteristics and memory properties, memristors are often used in neuromorphic circuits, particularly neural network circuits. However, most of the existing neural network circuits that can generate complex dynamics have high dimensions and excessive connections, which is not conducive to implementation. This article introduces a memristor containing an arctangent function into a simple cyclic neural network (SCNN) circuit to design a simple cyclic memristive neural network (SCMNN) circuit capable of generating complex multiscroll chaotic attractors. The designed SCMNN contains an external stimulus current and generates multiscroll attractors, with the number of scrolls expanding as the switches in the memristor equivalent circuit are activated. By varying the parameters, the multiscroll attractors can be broken into different numbers of coexisting attractors, which also depends on the switch, and it can achieve multivariable amplitude control when there is only one scroll. The anti-interference ability of the circuit is tested. A low-cost circuit-based microcontroller suitable for engineering applications is designed for it, and multiscroll attractors are successfully captured in an oscilloscope. The National Institute of Standards and Technology (NIST) test is carried out to verify its application value.

The augmented Lagrange Hopfield network for economic dispatch with multiple fuel options

A generalization of the energy function of the classical continuous Hopfield neural network is presented, the stationary points of which coincide with the complete set of equilibrium conditions of the network. Instead of applying statistical mechanical arguments, a direct proof is given. An energy expression of a Hopfield network having built-in constraints, namely of the so-called Potts glasses type, is presented. By performing a far-reaching generalization, the most general framework of continuous Hopfield networks is created, where almost arbitrary energy functions can be chosen and where constraints of all kind can be incorporated in the neural net. The analysis includes the presentation of several stability theorems concerning various sets of differential equations. Finally, a discussion on the possibilities to apply the presented theoretical results as well as an outlook on future topics of research is included.

New estimate on the domains of attraction of equilibrium points in continuous Hopfield neural networks

Focusing on features with higher importance degree: A significance-wise mechanism for image restoration

In image processing, Hopfield neural networks (HNNs) have been successively invested to solve the optimization model of image restoration. However, in the discrete case, HNNs lead to the fluctuation behaviors due to the limitation of the hard limit activators. The continuous model of HNNs generates non feasible solutions owing to the trial-and-error setting values process of the model parameters. To deal with these issues effectively, we suggest a new energy function with appropriate parameters settled by the stability analysis of the network. Performance of our method is demonstrated numerically and visually by several computational tests.

As a long-standing task, image restoration aims to recover the latent sharp image from its degraded counterpart. In recent years, owing to the strong ability of self-attention in capturing long-range dependencies, Transformer based methods have achieved promising performance on multifarious image restoration tasks. However, the canonical self-attention leads to quadratic complexity with respect to input size, hindering its further applications in image restoration. In this paper, we propose a Strip Attention Network (SANet) for image restoration to integrate information in a more efficient and effective manner. Specifically, a strip attention unit is proposed to harvest the contextual information for each pixel from its adjacent pixels in the same row or column. By employing this operation in different directions, each location can perceive information from an expanded region. Furthermore, we apply various receptive fields in different feature groups to enhance representation learning. Incorporating these designs into a U-shaped backbone, our SANet performs favorably against state-of-the-art algorithms on several image restoration tasks. The code is available at https://github.com/c-yn/SANet.

Recent works that utilized deep models have achieved superior results in various image restoration (IR) applications. Such approach is typically supervised, which requires a corpus of training images with distributions similar to the images to be recovered. On the other hand, the shallow methods, which are usually unsupervised remain promising performance in many inverse problems, e.g., image deblurring and image compressive sensing (CS), as they can effectively leverage nonlocal self-similarity priors of natural images. However, most of such methods are patch-based leading to the restored images with various artifacts due to naive patch aggregation in addition to the slow speed. Using either approach alone usually limits performance and generalizability in IR tasks. In this paper, we propose a joint low-rank and deep (LRD) image model, which contains a pair of triply complementary priors, namely, internal and external, shallow and deep, and non-local and local priors. We then propose a novel hybrid plug-and-play (H-PnP) framework based on the LRD model for IR. Following this, a simple yet effective algorithm is developed to solve the proposed H-PnP based IR problems. Extensive experimental results on several representative IR tasks, including image deblurring, image CS and image deblocking, demonstrate that the proposed H-PnP algorithm achieves favorable performance compared to many popular or state-of-the-art IR methods in terms of both objective and visual perception.

Compressed image restoration via deep deblocker driven unified framework

Memristor-based Hopfield network circuit for recognition and sequencing application

In image restoration tasks, like denoising and super resolution, continual modulation of restoration levels is of great importance for real-world applications, but has failed most of existing deep learning based image restoration methods. Learning from discrete and fixed restoration levels, deep models cannot be easily generalized to data of continuous and unseen levels. This topic is rarely touched in literature, due to the difficulty of modulating well-trained models with certain hyper-parameters. We make a step forward by proposing a unified CNN framework that consists of few additional parameters than a single-level model yet could handle arbitrary restoration levels between a start and an end level. The additional module, namely AdaFM layer, performs channel-wise feature modification, and can adapt a model to another restoration level with high accuracy. By simply tweaking an interpolation coefficient, the intermediate model - AdaFM-Net could generate smooth and continuous restoration effects without artifacts. Extensive experiments on three image restoration tasks demonstrate the effectiveness of both model training and modulation testing. Besides, we carefully investigate the properties of AdaFM layers, providing a detailed guidance on the usage of the proposed method.

Image restoration tasks such as deraining, deblurring, and dehazing are crucial for enhancing the environmental perception of autonomous vehicles and traffic systems, particularly for tasks like vehicle detection, pedestrian detection and lane line identification. While transformer-based models excel in these tasks, their prohibitive computational complexity hinders real-world deployment on resource-constrained platforms. To bridge this gap, this paper introduces a novel Soft Knowledge Distillation (SKD) framework, designed specifically for creating highly efficient yet powerful image restoration models. Our core innovation is twofold: first, we propose a Multi-dimensional Cross-Net Attention(MCA) mechanism that allows a compact student model to learn comprehensive attention relationships from a large teacher model across both spatial and channel dimensions, capturing fine-grained details essential for high-quality restoration. Second, we pioneer the use of a contrastive learning loss at the reconstruction level, treating the teacher’s outputs as positives and the degraded inputs as negatives, which significantly elevates the student’s reconstruction quality. Extensive experiments demonstrate that our method achieves a superior trade-off between performance and efficiency, notably enhancing downstream tasks like object detection. The primary contributions of this work lie in delivering a practical and compelling solution for real-time perceptual enhancement in autonomous systems, pushing the boundaries of efficient model design.

Image restoration aims to reconstruct a high-quality image from a degraded low-quality observation. Recently, Transformer models have achieved promising performance on image restoration tasks due to their powerful ability to model long-range dependencies. However, the quadratically growing complexity with respect to the input size makes them inapplicable to practical applications. In this paper, we develop an efficient convolutional network for image restoration by enhancing multi-scale representation learning. To this end, we propose an omni-kernel module that consists of three branches, i.e., global, large, and local branches, to learn global-to-local feature representations efficiently. Specifically, the global branch achieves a global perceptive field via the dual-domain channel attention and frequency-gated mechanism. Furthermore, to provide multi-grained receptive fields, the large branch is formulated via different shapes of depth-wise convolutions with unusually large kernel sizes. Moreover, we complement local information using a point-wise depth-wise convolution. Finally, the proposed network, dubbed OKNet, is established by inserting the omni-kernel module into the bottleneck position for efficiency. Extensive experiments demonstrate that our network achieves state-of-the-art performance on 11 benchmark datasets for three representative image restoration tasks, including image dehazing, image desnowing, and image defocus deblurring. The code is available at https://github.com/c-yn/OKNet.