A5013502609- Advanced Memory and Neural Computing
- Neuroscience and Neural Engineering
- Neural dynamics and brain function
- Neural Networks and Applications
- Smart Agriculture and AI
- Neural Networks Stability and Synchronization
- CCD and CMOS Imaging Sensors
- Advanced Vision and Imaging
- stochastic dynamics and bifurcation
- Cellular Automata and Applications
- Autonomous Vehicle Technology and Safety
- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- Photoreceptor and optogenetics research
- Robotics and Sensor-Based Localization
- Robotic Path Planning Algorithms
- Ferroelectric and Negative Capacitance Devices
- Image Processing Techniques and Applications
- Advanced Neural Network Applications
- Optical measurement and interference techniques
- Plant Disease Management Techniques
- AI in cancer detection
- Medical Imaging and Analysis
- Plant Virus Research Studies
- Dental Radiography and Imaging
Jeonbuk National University
2016-2025
Samsung Medical Center
2024
Sungkyunkwan University
2024
Hanyang University
2023
Sejong University
2023
Chonbuk National University Hospital
2004-2013
Korea Electrotechnology Research Institute
2012
Pusan National University
2012
Korea Automotive Technology Institute
2011
Chonnam National University
1997-2007
This paper illustrates that for a device to be memristor it should exhibit three characteristic fingerprints: 1) When driven by bipolar periodic signal the must "pinched hysteresis loop" in voltage-current plane, assuming response is periodic. 2) Starting from some critical frequency, lobe area decrease monotonically as excitation frequency increases, and 3) pinched loop shrink single-valued function when tends infinity. Examples of memristors exhibiting these fingerprints, along with...
A memristor emulator which imitates the behavior of a TiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> is presented. Our built from off-the-shelf solid state components. To develop real world circuit applications, can be used for breadboard experiments in time. Two or more emulators connected serial, parallel, hybrid (serial and parallel combined) with identical opposite polarities. With simple change connection, each switched between...
Analog hardware architecture of a memristor bridge synapse-based multilayer neural network and its learning scheme is proposed. The use synapse in the proposed solves one major problems, regarding nonvolatile weight storage analog implementations. To compensate for spatial nonuniformity nonideal response synapse, modified chip-in-the-loop suitable also In method, initial conducted software, behavior software-trained learned by each single-layered neurons independently. forward calculation...
A pulse-based programmable memristor circuit for implementing synaptic weights artificial neural networks is proposed. In the weighting circuit, both positive and negative multiplications are performed via a charge-dependent Ohm's law (). The composed of five memristors with bridge-like connections operates like an synapse processing adjustability. sign switching pulses, weight setting pulses applied through shared input terminal. Simulations done linear window-based nonlinear models.
This paper presents a rigorous and comprehensive nonlinear circuit-theoretic foundation for the memristive Hodgkin–Huxley Axon Circuit model. We show that comprises potassium ion-channel memristor sodium memristor, along with some mundane circuit elements. From this new perspective, many hitherto unresolved anomalous phenomena paradoxes reported in literature are explained clarified. The yet unknown dynamical mechanisms which give birth to action potentials remain hidden within memristors,...
In this paper, we propose a memristor bridge circuit consisting of four identical memristors that is able to perform zero, negative, and positive synaptic weightings. Together with three additional transistors, the weighting operation for neural cells. It compact as both weight programming are performed in synapse. power efficient, since based on pulsed input signals. Its terminals utilized commonly applying processing signals via time sharing. features synapses investigated using TiO model...
This exposition shows that the potassium ion-channels and sodium are distributed over entire length of axons our neurons in fact locally-active memristors. In particular, they exhibit all fingerprints memristors, including characteristic pinched hysteresis Lissajous figures voltage-current plane, whose loop areas shrink as frequency periodic excitation signal increases. Moreover, loops for ion-channel memristor, from Hodgkin-Huxley axon circuit model unique each signal. An in-depth...
Memristor-based circuit architecture for multilayer neural networks is proposed. It a first of its kind demonstrating successful circuit-based learning network built with memristors. Though back-propagation algorithm powerful scheme networks, hardware implementation very difficult due to complexities the synapses and operations involved in algorithm. In this paper, designed memristor bridge realized simple called Random Weight Change (RWC). RWC requires more iterations than algorithm, we...
This paper shows the action potential (spikes) generated from Hodgkin–Huxley equations emerges near edge of chaos consisting a tiny subset locally active regime HH equations. The main result proves that eigenvalues 4 × Jacobian matrix associated with mathematically intractable system four nonlinear differential are identical to zeros scalar complexity function theory. Moreover, we show loci pair complex-conjugate migrate continuously as an externally applied DC current excitation emulating...
To evaluate the diagnostic performance of a deep convolutional neural network (DCNN)-based computer-assisted diagnosis (CAD) system in detection osteoporosis on panoramic radiographs, through comparison with diagnoses made by oral and maxillofacial radiologists.Oral radiologists >10 years experience reviewed radiographs 1268 females {mean [± standard deviation (SD)] age: 52.5 ± 22.3 years} when cortical erosion mandibular inferior cortex was observed. Among females, 635 had no [mean (± SD)...
An attention guided convolutional neural network (CNN) for the classification of breast cancer histopathology images is proposed. Neural networks are generally applied as black box models and often network's decisions difficult to interpret. Making decision process transparent, hence reliable important a computer-assisted diagnosis (CAD) system. Moreover, it crucial that be based on histopathological features in agreement with human expert. To this end, we propose use additional region-level...
A method to utilize the memristor as a multilevel memory has been proposed. There are several roadblocks in practical use of memristors for memory. difficulty comes from nonlinearity ¿ vs. q curve which makes it difficult determine proper pulse width desired resistance values. Another one property integrates any kind signals including noise that appeared at and causes be perturbed their original The proposed enables used using reference array by forcing stick set predetermined fixed We...
Convolutional neural networks (CNNs) have achieved state-of-the-art performance in numerous aspects of human life and the agricultural sector is no exception. One main objectives deep learning for smart farming to identify precise location weeds crops on farmland. In this paper, we propose a semantic segmentation method based cascaded encoder-decoder network, namely CED-Net, differentiate from crops. The existing architectures are quite deep, with millions parameters that require longer...
This paper presents a novel vision based approach for detecting rows of crop in paddy field. The precise detection row enables farm-tractor to autonomously navigate the field successful inter-row weeding. While prior works on rely primarily various image features, deep neural network learning semantic graphics directly extract from an input is used this work. A convolutional encoder decoder trained detect lines using graphics. detected are then derive control signal steering tractor results...
Nuclei segmentation and classification of hematoxylin eosin-stained histology images is a challenging task due to variety issues, such as color inconsistency that results from the non-uniform manual staining operations, clustering nuclei, blurry overlapping nuclei boundaries. Existing approaches involve segmenting by drawing their polygon representations or measuring distances between centroids. In contrast, we leverage fact morphological features (appearance, shape, texture) in tissue vary...
Three-phase current measurement using only a single sensor in the inverter dc link is appealing for ac motor drives because it minimizes number of sensors, thereby reducing cost, weight, and volume. However, basic technique poses special challenges duration active voltage vectors must be long enough to measure reliably during every pulsewidth modulation switching interval. A new algorithm reconstructing phase currents called vector insertion method (MVIM) presented that overcomes this...
In this paper, a generic model of memristive systems, which can emulate the behavior real devices is proposed. Non-ideal pinched hysteresis loops are sometimes observed in devices. For example, may deviate from origin over broad range amplitude A and frequency f input signal. This deviation ideal case often caused by parasitic circuit elements exhibited we propose adding four elements, namely, small capacitance, inductance, DC current source, voltage to device. The adequacy verified...
A memristor bridge neural circuit which is able to perform signed synaptic weighting was proposed in our previous study, where the operation verified via software simulation of mathematical model HP memristor. This study an extension work advancing toward implementation architecture synapse built with emulator circuits. In addition, a simple network performs both and summation by combining emulators-based synapses differential amplifier The feasibility SPICE simulations.
This paper demonstrates that the Chua Corsage Memristor, when connected in series with an inductor and a battery, oscillates about locally-active operating point located on memristor’s DC [Formula: see text]–[Formula: text] curve. On point, small-signal equivalent circuit is derived via Taylor expansion. The admittance from value of inductance determined at frequency where real part Memristor zero. Oscillation analyzed in-depth application theory Local Activity, Edge Chaos Hopf-bifurcation.
Abstract Memristors represent the fourth electrical circuit element complementing resistors, capacitors and inductors. Hallmarks of memristive behavior include pinched frequency-dependent I–V hysteresis loops most importantly a functional dependence magnetic flux passing through an ideal memristor on its charge. Microtubules (MTs), cylindrical protein polymers composed tubulin dimers are key components cytoskeleton. They have been shown to increase solution’s ionic conductance re-orient in...
In the domain of medical image analysis, deep learning models are heralding a revolution, especially in detecting complex and nuanced features characteristic diseases like tumors cancers. However, robustness adaptability these across varied imaging conditions magnifications remain formidable challenge. This paper introduces Fourier Adaptive Recognition System (FARS), pioneering model primarily engineered to address malarial parasite recognition. Yet, foundational principles guiding FARS lend...