• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.436-439
• /
• 2007

Noise reduction in the image is very important to improve the quality of the image. This paper discusses a new noise reduction technique which uses the On/Off spatio-temporal structure of the receptive fields. Also this paper proposes a structurally improved artificial vision system which incorporates the sphere type retina structure, an improved On/Off spatio-temporal receptive fields structure, and chiasm for hemianopia testing.

• Progress in Medical Physics
• /
• v.17 no.4
• /
• pp.260-267
• /
• 2006

A retinal ganglion cell's receptive field is defined as that region on the retinal surface In which a light stimulus will produce a response. A retinal ganglion cell peers out at a small patch of the visual scene through its receptive field and encodes local features with action potentials that pass through the optic nerve to higher centers. Therefore, defining the receptive field of a retinal ganglion cell is essential to understand the electrical characteristics of a ganglion cell. Distribution of receptive fields over retinal surface provides us an Insight how the retinal ganglion cell processes the visual scene. In this paper, we provide the details how to reconstruct the receptive field of a retinal ganglion cell. We recorded the ganglion cell's action potential with multielectrode array when the random checkerboard stimulus was applied. After classifying the retinal waveform Into ON-cell, OFF-cell, ON/OFF-cell, we reconstructed the receptive field of retinal ganglion cell with Matlab. Here, we show the receptive fields of ON-cell and OFF-cell.

• The Korean Journal of Physiology and Pharmacology
• /
• v.2 no.1
• /
• pp.21-30
• /
• 1998

The effects of GABA and glutamate on the horizontal cells were explored by an intracellular recording method to discern the mechanisms of receptive field formation by chemical coupling in the catfish outer retina. The results suggest that the horizontal cells of the catfish retina might use GABA as their transmitters and that the GABAergic system contributes to the formation of receptive fields of the horizontal cells. GABAC receptors may be involved in a chemical coupling between horizontal cells and concerned with the depolarizing actions by GABA on horizontal cells in the catfish retina. Since the chloride equilibrium potential is more positive than the dark membrane potential in horizontal cells, GABA released from a horizontal cell may depolarize the neighboring horizontal cells. Thus a chemical coupling between horizontal cells may be formed. $GABA_A$ receptors also may be involved in the negative feedback mechanism between photoreceptor and horizontal cell. And glutamate may be involved in connecting positive and negative feedback systems since it potentiated the GABA's actions. Therefore, it is presumed that large receptive fields in the catfish retina are formed not only by electrical coupling but also by chemical coupling between horizontal cells. And information travels laterally by pathways involving both electrical coupling composed of gap junctions and chemical coupling in the retinal network.

• Journal of Biomedical Engineering Research
• /
• v.16 no.1
• /
• pp.33-42
• /
• 1995

A new neural network architecture for the recognition of patterns from images is proposed, which is partially based on the results of physiological studies. The proposed network is composed of multi-layers and the nerve cells in each layer are connected by spatial filters which approximate receptive fields in optic nerve fields. In the proposed method, patterns recognition for complicated images is carried out using global features as well as local features such as lines and end-points. A new generating method of matched filters representing global features is proposed in this network.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.620-623
• /
• 2003

The TAM (Topographic Attentive Mapping) network is a biologically-motivated neural network. Fuzzy rules are acquired from the TAM network by the pruning algorithm. In this paper we formulate a new input layer using Gabor function for TAU network to realize receptive field of human visual cortex.

• The Journal of the Korea Contents Association
• /
• v.5 no.6
• /
• pp.87-96
• /
• 2005

In this paper, a selective visual attention module based on motion stimuli is introduced for the purpose of detecting ROI(region of interest) or FOA(focus of attention) in motion pictures. Analysis of motion fields in our approach is in direct contrast to some of the previous studies of selective visual attention module. Motion that presents temporal visual saliency in an aspect between two successive frames is analyzed based on psychological studies in 'DORF(double opponent receptive fields)' and 'NF(noise filtration)' in MT(middle temporal cortex). Analyzed results are integrated based on the theory of 'motion integration' in MT to obtain a single conspicuous region. Experiments through a human subjective evaluation showed generally accepted results.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.319-325
• /
• 2022

The variation of utterance lengths is a representative factor that can degrade the performance of speaker verification systems. To handle this issue, previous studies had attempted to extract speaker features from various branches or to use convolution layers with different receptive fields. Combining the advantages of the previous two approaches for variable-length input, this paper proposes integrated receptive field diversification that extracts speaker features through more diverse receptive field. The proposed method processes the input features by convolutional layers with different receptive fields at multiple time-axis branches, and extracts speaker embedding by dynamically aggregating the processed features according to the lengths of input utterances. The deep neural networks in this study were trained on the VoxCeleb2 dataset and tested on the VoxCeleb1 evaluation dataset that divided into 1 s, 2 s, 5 s, and full-length. Experimental results demonstrated that the proposed method reduces the equal error rate by 19.7 % compared to the baseline.

• Journal of Korea Multimedia Society
• /
• v.24 no.1
• /
• pp.1-12
• /
• 2021

In this paper, we present a new architecture for semantic segmentation. Semantic segmentation aims at a pixel-wise classification which is important to fully understand images. Previous semantic segmentation networks use features of multi-layers in the encoder to predict final results. However, they do not contain various receptive fields in the multi-layers features, which easily lead to inaccurate results for boundaries between different classes and small objects. To solve this problem, we propose a multi-path feature fusion module that allows for features of each layers to contain various receptive fields by use of a set of dilated convolutions with different dilatation rates. Various experiments demonstrate that our method outperforms previous methods in terms of mean intersection over unit (mIoU).

• The Korean Journal of Physiology
• /
• v.24 no.2
• /
• pp.389-402
• /
• 1990

The physiological characteristics of the neurons receiving the ventral root afferent inputs were investigated in the cat. A total of 70 cells were identified in the lumbosacral spinal cord. All these cells responded only to the C-strength stimulation of the distal stump of cut ventral root and the estimated conduction velocities of the VRA fibers were not faster than 4 m/sec. The majority of them were silent in resting state. For 49 cells, their peripheral receptive fields were characterized. Among them, 25 cells were exclusively excited by VRA inputs, 8 were inhibited and the remaining cells recevied both excitatory and inhibitory VRA inputs. According to the response pattern to the mechanical stimuli applied to their receptive fields, only a fourth of them were typical high threshold cell, a sixth, wide dynamic range cells, while remainings were a rather complex cells. Most of the cells receiving VRA inputs, received only the A ${\delta}-peripheral$ nerve inputs. Intravenous injection of morphine decreased the response of spinal cells to the VRA activation. The responses were abolished completely by counter irritation to the common peroneal nerve with C-strength-low frequency stimuli. These physiological properties of the spinal neurons receiving the VRA inputs are differ in some aspect from the spinal neurons receiving nociceptive inputs from the periphery, but still were consistent with the contention that VRA system might carry nociceptive informations arising from the spinal cord and/or neraby surrounding tissues.

• The Korean Journal of Physiology
• /
• v.28 no.2
• /
• pp.113-122
• /
• 1994

The present study was carried out to characterize the functional properties of spinomesencephalic tract (SMT) neurons in the lumbar spinal cord of urethane anesthetized rats. Extracellular single unit recordings were made from neurons antidromically activated by stimulation of the midbrain area, including the deep layers of superior colliculus, periaqueductal gray and midbrain reticular formation. Recording sites were located in laminae I-VII of spinal cord segments of L2-L5. Receptive field properties and responses to calibrated mechanical stimulation were studied in 78 SMT cells. Mean conduction velocity of SMT neurons was $19.1{\pm}1.04\;m/sec$. SMT units were classified according to their response profiles into four groups: wide dynamic range (58%), deep/tap (23%), high threshold (9%) and low threshold (3%). A simple excitatory receptive field was found for most SMT neurons recorded in superficial dorsal horn (SDH). Large complex inhibitory and/or excitatory receptive fields were found for cells in lateral reticulated area which usually showed long after-discharge. Most of SMT cells received inputs from $A{\delta}$ and C afferent fiber types. These results suggest that sensory neurons in the rat SMT may have different functional roles according to their location in the spinal cord in integrating and processing sensory inputs including noxious mechanical stimuli.