• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.808-810
• /
• 2018

Pedestrian tracking is a particular problem of object tracking, and an important component in various vision-based applications, such as autonomous cars or surveillance systems. After several years of development, pedestrian tracking in videos is still a challenging problem because of various visual properties of objects and surrounding environment. In this research, we propose a tracking-by-detection system for pedestrian tracking, which incorporates Convolutional Neural Network (CNN) and color information. Pedestrians in video frames are localized by a CNN, then detected pedestrians are assigned to their corresponding tracklets based on similarities in color distributions. The experimental results show that our system was able to overcome various difficulties to produce highly accurate tracking results.

• Journal of Information Processing Systems
• /
• v.15 no.1
• /
• pp.67-90
• /
• 2019

Accurate detection, tracking and analysis of human movement using robots and other visual surveillance systems is still a challenge. Efforts are on to make the system robust against constraints such as variation in shape, size, pose and occlusion. Traditional methods of detection used the sliding window approach which involved scanning of various sizes of windows across an image. This paper concentrates on employing a state-of-the-art, hierarchical graph based method for segmentation. It has two stages: part level segmentation for color-consistent segments and object level segmentation for category-consistent regions. The tracking phase is achieved by employing SIFT keypoint descriptor based technique in a combined matching and tracking scheme with validation phase. Localization of human region in each frame is performed by keypoints by casting votes for the center of the human detected region. As it is difficult to avoid incorrect keypoints, a consensus-based framework is used to detect voting behavior. The designed methodology is tested on the video sequences having 3 to 4 persons.

• International Journal of Advanced Culture Technology
• /
• v.9 no.4
• /
• pp.1-8
• /
• 2021

Since the late 20th century, various projects in the public domain where local communities and art organizations collaborate have been attempted in miscellaneous ways. In terms of anticipating an active attitude of the community, socially engaged project focuses on proactively changing everyday life and environment of people. An art collective, Assemble who won the Turner Prize in 2015 for one of their projects named Granby Four Streets (2013) has been admitted as an exemplary of this phenomenon, and it appears frequently in discussions about community-led projects. We investigate Granby Four Streets, as well as the documentation formed by the perspective of third parties based on the surveillance and criticism aspects. It asserts that the limitation of socially engaged projects that are kept ephemerally and easily forgotten could be overcome with the concept of documentation and its practice.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.3
• /
• pp.60-66
• /
• 2012

In this paper, we propose abandoned object detection algorithm. When abandoned object was occluded other object, the existing methods cannot detect abandoned object because those methods are not able to estimate the location of abandoned object. In order to overcome this problem, the proposed algorithm extracts the corners around abandoned object. The detected corners are linked to center of abandoned object called by supporters. We can then estimate the location of abandoned object by using supporters. Therefore, the proposed algorithm can detect and estimate the location of abandoned object, when abandoned object is occluded by other object. For this reason, the proposed algorithm can be applied to intelligent surveillance system to prevent bomb terror, which disguises as luggage or box.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.27-33
• /
• 2020

Recently, visual object detection and tracking has become a vital role in many different applications. It spans various applications like robotics, video surveillance, and intelligent vehicle navigation. Especially, in current situation where the use of UAVs is expanding widely, detection and tracking to soot down illegal UAVs flying over the sky at airports, nuclear power plants and core facilities is becoming a very important task. The remarkable method in object tracking is correlation filter based tracker like KCF (Kernelized Correlation Filter). But it has problems related to target drift in tracking process for long-term tracking. To mitigate the target drift problem in video surveillance application, we propose a tracking method which uses KCF, adaptive thresholding and Kalman filter. In the experiment, the proposed method was verified by using monochrome video sequences which were obtained in the operational environment of UAV.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.4
• /
• pp.787-794
• /
• 2018

The various natural disasters such as floods and localized heavy rains are increasing due to the global warming. If a natural disaster can be detected and analyzed in advance and more effectively, it can prevent enormous damage of natural disasters. Recent development in visual sensor technologies has encouraged various studies on monitoring environments including rivers. In this paper, we propose a method to detect water regions from river images which can be exploited for river surveillance systems using video sensor networks. In the proposed method, we first segment a river image finely using the minimum spanning tree algorithm. Then, the seed regions for the river region and the background region are set by using the preliminary information, and each seed region is expanded by merging similar regions to segment the water region from the image. Experimental results show that the proposed method separates the water region from a river image easier and accurately.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.557-561
• /
• 2023

This study aims to assess how effective environmental management can be accomplished through monitoring of environmental conditions in patient discharge rooms within healthcare facilities through direct observation and the use of fluorescent markers. From March to July 2013, this study evaluated 448 check-out beds in wards and intensive care units before and 494 after intensive environmental monitoring activities. The collected data were analyzed using the SPSS 21.0 program. According to the study's findings, direct observation increased from 95.2% prior to the implementation of intensive environmental monitoring activities to 98.9% following the implementation, which was statistically significant. The non-detection rate of fluorescent markers exhibited an increase from 96.1% prior to the commencement of intensive environmental monitoring activities to 98.0% following their implementation. However, it should be noted that this observed increase was not deemed statistically significant. In light of the results of this research, it is imperative to evaluate the effectiveness of environmental management by employing a variety of assessment methods, including direct observation and fluorescent markers.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.2
• /
• pp.719-722
• /
• 2016

Background: Cholangiocarcinoma (CCA) is a serious health problem in Thailand, particularly in northeastern and northern regions, but epidemiological studies are scarce and the spatial distribution of CCA remains to be determined. A database for the population at risk is required for monitoring, surveillance and organization of home health care. This study aim was to geo-visually display the distribution of CCA in northeast Thailand, using a geographic information system and Google Earth. Materials and Methods: A cross-sectional survey was carried out in 9 sub-districts and 133 villages in Chum Phuang district, Nakhon Ratchasima province during June and October 2015. Data on demography, and the population at risk for CCA were combined with the points of villages, sub-district boundaries, district boundaries, and points of hospitals in districts, then fed into a geographical information system. After the conversion, all of the data were imported into Google Earth for geo-visualization. Results: A total of 11,960 from 83,096 population were included in this study. Females and male were 52.5%, and 47.8%, the age group 41-50 years old 33.3%. Individual risk for CCA was identifed and classified by using the Korat CCA verbal screening test as low (92.8%), followed by high risk (6.74%), and no (0.49%), respectively. Gender ($X^2$-test=1143.63, p-value= 0.001), age group ($X^2$-test==211.36, p-value=0.0001), and sub-district ($X^2$-test=1471.858, p-value=0.0001) were significantly associated with CCA risk. Spatial distribution of the population at risk for CCA in Chum Phuang district was viewed with Google Earth. Geo-visual display followed Layer 1: District, Layer 2: Sub-district, Layer 3: Number of low risk in village, Layer 4: Number of high risk in village, and Layer 5: Hospital in Chum Phuang District and their related catchment areas. Conclusions: We present the first risk geo-visual display of CCA in this rural community, which is important for spatial targeting of control efforts. Risk appears to be strongly associated with gender, age group, and sub-district. Therefor, spatial distribution is suitable for the use in the further monitoring, surveillance, and home health care for CCA.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.3
• /
• pp.1293-1297
• /
• 2016

Cholangiocarcinoma (CCA) is a serious problem in Thailand, particularly in the northeastern and northern regions. Database of population at risk are need required for monitoring, surveillance, home health care, and home visit. Therefore, this study aimed to develop a geographic information system (GIS) database and Google map of the population at risk of CCA in Mueang Yang district, Nakhon Ratchasima province, northeastern Thailand during June to October 2015. Populations at risk were screened using the Korat CCA verbal screening test (KCVST). Software included Microsoft Excel, ArcGIS, and Google Maps. The secondary data included the point of villages, sub-district boundaries, district boundaries, point of hospital in Mueang Yang district, used for created the spatial databese. The populations at risk for CCA and opisthorchiasis were used to create an arttribute database. Data were tranfered to WGS84 UTM ZONE 48. After the conversion, all of the data were imported into Google Earth using online web pages www.earthpoint.us. Some 222 from a 4,800 population at risk for CCA constituted a high risk group. Geo-visual display available at following www.google.com/maps/d/u/0/edit?mid=zPxtcHv_iDLo.kvPpxl5mAs90&hl=th. Geo-visual display 5 layers including: layer 1, village location and number of the population at risk for CCA; layer 2, sub-district health promotion hospital in Mueang Yang district and number of opisthorchiasis; layer 3, sub-district district and the number of population at risk for CCA; layer 4, district hospital and the number of population at risk for CCA and number of opisthorchiasis; and layer 5, district and the number of population at risk for CCA and number of opisthorchiasis. This GIS database and Google map production process is suitable for further monitoring, surveillance, and home health care for CCA sufferers.

• Korean Journal of Remote Sensing
• /
• v.36 no.2_2
• /
• pp.365-377
• /
• 2020

Continuous monitoring and immediate response is essential to protect the national maritime territory and maritime resources from the activities of illegal ships. Synthetic Aperture Radar (SAR) images with a wide range of images are effective for maritime surveillance asthe weather and day-night conditions rarely affect to image acquisition. However, an effective ship detection is not easy due to the huge data size of SAR images and various characteristics such as the speckle noise. In this study, the Human Visual Attention System (HVAS) algorithm was applied to KOMPSAT-5 to extract the initial targets, and the SAR-Split algorithm depending on the imaging modes was used to remove false alarms. The detected targets were finally selected by the Constant False Alarm Rate (CFAR) algorithm and matched with the ship's Automatic Identification System (AIS) information. Overall, the detected targets were well matched with AIS data, but some false alarms by ship wakes were observed. The detection rate was about 80% in ES mode and about 64% in ST mode. It is expected that the developed ship detection algorithm will contribute to the construction of a wide area maritime surveillance network.