• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.5
• /
• pp.327-334
• /
• 1991

This paper proposes an efficient method of implementing the generalized Hough transform (GHT), which has been hindered by an excessive computing load and a large memory requirement. The conventional algorithm requires a parameter space of 4 dimensions in detection a rotated, scaled, and translated object in an input image. Prior to the application of GHT to the input image, the proposed method determines the angle of rotation and the scaling factor of the test image using the proportion of the edge components between the reference image and test image. With the rotation angle and the scaling factor already determined, the parameter spaceis to be reduced to a simple array of 2 dimensions by applying the unit GHT only one time. The experiments with the image of airplanes reveal that both of the computing time and the requires memory size are reduced by 95 percent, without any degradatationof accuracy, compared with the conventional GHT algorithm.

• Journal of IKEEE
• /
• v.17 no.2
• /
• pp.202-207
• /
• 2013

This paper implemented the conventional FAST and BRIEF algorithm as hardware on Zynq-7000 SoC Platform. Previous feature-based hardware accelerator is mostly implemented using the SIFT or SURF algorithm, but it requires excessive internal memory and hardware cost. The proposed FAST & BRIEF accelerator reduces approximately 57% of internal memory usage and 70% of hardware cost compared to the conventional SIFT or SURF accelerator, and it processes 0.17 pixel per Clock.

• ETRI Journal
• /
• v.39 no.4
• /
• pp.502-513
• /
• 2017

For video understanding, namely analyzing who did what in a video, actions along with objects are primary elements. Most studies on actions have handled recognition problems for a well-trimmed video and focused on enhancing their classification performance. However, action detection, including localization as well as recognition, is required because, in general, actions intersect in time and space. In addition, most studies have not considered extensibility for a newly added action that has been previously trained. Therefore, proposed in this paper is an extensible hierarchical method for detecting generic actions, which combine object movements and spatial relations between two objects, and inherited actions, which are determined by the related objects through an ontology and rule based methodology. The hierarchical design of the method enables it to detect any interactive actions based on the spatial relations between two objects. The method using object information achieves an F-measure of 90.27%. Moreover, this paper describes the extensibility of the method for a new action contained in a video from a video domain that is different from the dataset used.

• Korean Journal of Pharmacognosy
• /
• v.50 no.4
• /
• pp.253-259
• /
• 2019

Chaenomeles speciose Nakai (CSP) or Chaenomeles sinensis Koehne (CSS) (Rosaceae) has been used, traditionally, to treat muscle problems and gastric dampness in eastern Asia countries. Therefore, many studies have focused on investigating its active compounds and effects on muscle pain, arthritis and gastro-intestinal diseases. Recently, several studies reported that CSS extract degrade amyloid plaques and enhance synaptic acetylcholine level in vivo and in vitro. Although these two Chaenomeles species are used without differences, CSP is reported to contains more phenolic compounds which are known to enhance memory. Therefore, in this study, we investigated the memory ameliorating effects of CSP by employing the passive avoidance test, Y-maze task and novel object recognition test. CSP (30 or 100 mg/kg) ameliorated the declined memory induced by scopolamine injection and enhanced the brain-derived neurotrophic factor (BDNF) levels along with post synaptic density protein 95 (PSD 95) levels at the hippocampus of the scopolamine-injected mouse brain. These results suggested that CSP alleviates the cognition declines caused by cholinergic blockade via enhancing BDNF levels and PSD 95, and that it would enhance memory formation and be useful for treating memory declines.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Anatomy and Cell Biology
• /
• v.56 no.4
• /
• pp.494-507
• /
• 2023

Vascular dementia (VaD) is characterized by progressive memory impairment, which is associated with microglia-mediated neuroinflammation. Polyphenol-rich natural plants, which possess anti-inflammatory activities, have attracted scientific interest worldwide. This study investigated whether Rubus fruticosus leaf extract (RFLE) can attenuate VaD. Sprague-Dawley rats were separated into five groups: SO, sham-operated and treated with vehicle; OP, operated and treated with vehicle; RFLE-L, operated and treated with low dose (30 mg/kg) of RFLE; RFLE-M, operated and treated with medium dose (60 mg/kg) of RFLE; and RFLE-H, operated and treated with high dose (90 mg/kg) of RFLE. Bilateral common carotid artery and hypotension were used as a modeling procedure, and the RFLE were intraorally administered for 5 days (preoperative 2 and postoperative 3 days). The rats then underwent memory tests including the novel object recognition, Y-maze, Barnes maze, and passive avoidance tests, and neuronal viability and neuroinflammation were quantified in their hippocampi. The results showed that the OP group exhibited VaD-associated memory deficits, neuronal death, and microglial activation in hippocampi, while the RFLE-treated groups showed significant attenuation in all above parameters. Next, using BV-2 microglial cells challenged with lipopolysaccharide (LPS), we evaluated the effects of RFLE in dynamics of proinflammatory mediators and the upstream signaling pathway. RFLE pretreatment significantly inhibited the LPS-induced release of nitric oxide, TNF-α, and IL-6 and upregulation of the MAPKs/NF-κB/iNOS pathway. Collectively, we suggest that RFLE can attenuate the histologic alterations and memory deficits accompanied by VaD, and these roles are, partly due to the attenuation of microglial activation.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.1
• /
• pp.97-104
• /
• 2012

Natural User Interface(NUI) technologies introduce new trends in using devices such as computer and any other electronic devices. In this paper, an augmented reality on a mobile device is implemented by using face detection, recognition and motion tracking. The face detection is obtained by using Viola-Jones algorithm from the images of the front camera. The Eigenface algorithm is employed for face recognition and face motion tracking. The augmented reality is implemented by overlapping the rear camera image and GPS, accelerator sensors' data with the 3D graphic object which is correspond with the recognized face. The algorithms and methods are limited by the mobile device specification such as processing ability and main memory capacity.

• Korean Journal of Cognitive Science
• /
• v.6 no.2
• /
• pp.97-116
• /
• 1995

This study was conducted to investigate how syntaction and pragmatic constraints influence the sentential representation and memory accessibility. In order to seperate the syntactic constraints from the pragmatic constraint from the pragmatic constraints,the syntactic role of constituent in the sentence (subject or object) and the order of mention(first or second) were manipulted.After each sentence was presented by RSVP procedure,the probe recognition time was measured to investigate memory accessibility.In Experiment 1,in which SOA interval was 255ms,it was found that the subject of a sentece were more accessible than the object and participants first in a sentence were more accessible than participants mentioned later.However, in Experiment 2,in which SOA interval was 1540ms,it was found that participants mentioned first in a sentence were more accessible than participants mentioned later while there was no significant difference between the subject and object of a sentece.These results suggest that the syntactic and pragmatic constraints have an independent effect on the initial senential representation at the early stage of constructing representation,but as time passes only the pragmatic constraints influence sentential representation.These results also support a theoretical position which assumes that sentential representation is constructed through the process of convergent statisfaction of multiple constraints.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.4
• /
• pp.61-68
• /
• 2007

Two experiments examined the visual memory distortion by presenting 170 subjects with wide-angle views of four scenes on monitor screen. Retention interval of 0, 1 and 48 hours tested in reproduction and recognition experiment. The results of reproduction showed that the subjects tend to magnify the foreground and background of scenes compared to the real input (scene) for all retention intervals. The viewers recognized mure wide-angle views for the same scenes at the retention interval of 1 and 48 hours. These results demonstrated boundary extension is not a robust and unidirectional phenomenon but boundary contraction can be occurred with wide-angle views. The results also suggested that boundary contraction is the product of the activation of a memory schema hypothesis: In memory the representation moves toward a prototypical view and prototypical object size.

• Korean Journal of Medicinal Crop Science
• /
• v.28 no.2
• /
• pp.85-94
• /
• 2020

Background: Alzheimer's disease (AD) is caused by various factors, such as cholinergic dysfunction, regulation of neurotrophic factor expression, and accumulation of amyloid-beta. We investigated whether or not a combination of Carthamus tinctorius L. seed and Taraxacum coreanum (CT) has a protective effect on scopolamine-induced memory impairment in a mouse model. Methods and Results: Mice were orally pretreated with CT (50, 100 and 200 mg/kg/day) for 14 days, and scopolamine (1 mg/kg/day) was injected intraperitoneally before subjecting them to behavior tests. CT-administered mice showed better novel object recognition and working memory ability than scopolamine-treated control mice. In T-maze and Morris water maze tests, CT (100 and 200 mg/kg/day) significantly increased space perceptive ability and occupancy to the target quadrant, respectively. In addition, 100 and 200 mg/kg/day of CT attenuated cholinergic dysfunction through inhibition of butyryl cholinesterase in brain tissue. Furthermore, CT-administered mice showed higher cyclic adenosine monophosphate-response element-binding protein (CREB) levels and lower amyloid precursor protein (APP) levels compared to scopolamine-treated control mice. Conclusions: CT improved scopolamine-induced memory impairment through inhibition of cholinergic dysfunction, up-regulation of CREB, and down-regulation of APP. Therefore, CT could be a useful therapeutic agent for AD with protective effects on cognitive impairment.