• 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.

• Convergence Security Journal
• /
• v.8 no.4
• /
• pp.97-104
• /
• 2008

Recently, the wireless positioning techniques and mobile computing techniques have rapidly developed to use location data of moving objects. The more the number of moving objects is numerous and the more periodical sampling of locations is frequent, the more location data of moving objects become very large. Hence the system should be able to efficiently manage mass location data, support various spatio-temporal queries for LBS, and solve the uncertainty problem of moving objects. Therefore, in this paper, innovating the location data of moving object effectively, we propose Rend 3DR-tree method to decrease the dead space and complement the overlapping of nodes by utilizing 3DR-tree with the indexing structure to support indexing of current data and history data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4795-4815
• /
• 2020

Drone detection can be considered as a specific sort of small object detection, which has always been a challenge because of its small size and few features. For improving the detection rate of drones, we design a Deeper SSD network, which uses large-scale input image and deeper convolutional network to obtain more features that benefit small object classification. At the same time, in order to improve object classification performance, we implemented the up-sampling modules to increase the number of features for the low-level feature map. In addition, in order to improve object location performance, we adopted the down-sampling modules so that the context information can be used by the high-level feature map directly. Our proposed Deeper SSD and its variants are successfully applied to the self-designed drone datasets. Our experiments demonstrate the effectiveness of the Deeper SSD and its variants, which are useful to small drone's detection and recognition. These proposed methods can also detect small and large objects simultaneously.

• Journal of Korea Spatial Information System Society
• /
• v.10 no.3
• /
• pp.67-78
• /
• 2008

With the rapid development of wireless communication and mobile equipment, many applications for location-based services have been emerging. Moving objects such as vehicles and ships change their positions over time. Moving objects have their moving path, called the trajectory, because they move continuously. To monitor the trajectory of moving objects in a large scale database system, an efficient Indexing scheme to processed queries related to trajectories is required. In this paper, we focus on the issues of minimizing the dead space of index structures. The Minimum Bounding Boxes (MBBs) of non-leaf nodes in trajectory-preserving indexing schemes have large amounts of dead space since trajectory preservation is achieved at the sacrifice of the spatial locality of trajectories. In this thesis, we propose entry relocating techniques to reduce dead space and overlaps in non-leaf nodes. we present performance studies that compare the proposed index schemes with the TB-tree and the R*-tree under a varying set of spatio-temporal queries.

• Journal of Korea Multimedia Society
• /
• v.11 no.2
• /
• pp.228-236
• /
• 2008

Recently augmented reality technology has been researched to view the virtual shape of buildings before construction or cultural heritages under recovery. Those researches used special devices or markers that are not applicable in long distanced outdoor environment. Also the server had to compute a lot of transformations for the location changes of virtual objects. This paper proposed a mobile augmented reality system that uses GPS and accelerometer sensors in order to compute the virtual object's locations without using markers. The server determines the position and orientation by comparing the GPS data obtained from the client with the predefined 3D object informations in the server. If the server sends the virtual object informations such as the position, orientation and matching information, then the client matches the virtual object on the screen of mobile camera phone. In addition, the client computes the transformations of location change detected by the accelerometer derived from the user's movement without additional connection to the server.

• Journal of Korea Spatial Information System Society
• /
• v.5 no.1 s.9
• /
• pp.65-74
• /
• 2003

Location-based service is an important spatiotemporal database application area that provides the location-aware information of wireless terminals via positioning devices such as GPS. With the rapid advances of wireless communication systems, the requirement of mobile application areas including traffic, mobile commerce and supply chaining management became the center of attention for various research issues in spatiotemporal databases. In this paper we present the A-Quadtree, an efficient indexing method for answering location-based queries where the movement vector information (e.g., speed and velocity) is not presented. We implement the A-Quadtree with an index structure for object identifiers as a.Net component to apply the component to multiplatforms. We present our approach and describe the performance evaluation through various experiments. In our experiments, we compare the performance with previous approaches and show the enhanced efficiency of our method.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.8 no.6
• /
• pp.26-34
• /
• 1999

This study shows an alternative method for the determination of object's position, based on a computer vision method. This approach develops the vision system model to define the reciprocal relationship between the 3-D real space and 2-D image plane. The developed model involves the bilinear six-view parameters, which is estimated using the relationship between the camera space location and real coordinates of known position. Based on estimated parameters in independent cameras, the position of unknown object is accomplished using a sequential estimation scheme that permits data of unknown points in each of the 2-D image plane of cameras. This vision control methods the robust and reliable, which overcomes the difficulties of the conventional research such as precise calibration of the vision sensor, exact kinematic modeling of the robot, and correct knowledge of the relative positions and orientation of the robot and CCD camera. Finally, the developed vision control method is tested experimentally by performing determination of object position in the space using computer vision system. These results show the presented method is precise and compatible.

• Journal of the Korean Society of Costume
• /
• v.64 no.3
• /
• pp.1-12
• /
• 2014

The purpose of this study is to investigate the Depaysement techniques in a new perspective, which are applicable to fashion window display. It was investigated by studying the case of Depaysement expressed in contemporary fashion window display. The analysis object of this study was limited to window displays shown at the world's most famous department stores in the last five years. The data was collected through related specialty publications and each department store's websites. The framework for analysis of this study is established by relevant precedent studies. The results of this study were drawn form comparative quantitative analysis from an expert group. Through the study, the characteristics of Depaysement in the contemporary fashion window display were classified into 'Change of forms and materials', 'Heterogeneous combination of objects', 'Location change of an object', 'Conversion of recognition on an object' and 'Change of spatial awareness'. The expression approaches were 'Change of scale', 'Change of materials', 'Combination of heterogeneous objects', 'Heterogeneous combination', 'Arrangement of object in a strange space', 'Change of display method', 'Overlapped object', 'Paradoxical image', 'Variable awareness of boundary' and 'Reorganization of interior space and change of materials'.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.21 no.45
• /
• pp.33-45
• /
• 1998

The typical characteristic of PDM(Product Data Management) System seperates the databases to store the meta data and applications. Therefore, meta data contains the information for the location of file, user profiles, relationships between the files, and process. PDM utilizes these information efficiently and does file management, configuration management, and process management. In this view, the integration strategy of PDM is to merge data and process. In the view of architecture, the interface between data and application and the actions of each application execute seamlessly. This architecture is viewed as integrated data and process among enterprises and implemented with client/server technology in distributed process environment that interfaced with open object-oriented technology which is developed with business object in the object-oriented infrastructure. In this paper, we studied the definition, function, and scope of PDM and researched the core technologies to implement the PDM integration environment. We also researched the PDM utilization in distributed enterprise environment and implementation of PDM integration environment with this technical background.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.1
• /
• pp.110-116
• /
• 2013

With the recent development of wireless network technologies, there have been increasing usage of variouse position base servies. Position based services basically collect position information of moving object for the utilization of them in real life. Accordingly, new index structures are required to efficiently retrieve the consecutive positions of moving objects. In the paper, we consider volume of Extended Minimum Bounding Rectangles(EMBR) to be determined by average size of range queries. We proposed the methode that split efficiently moving object with long distance between location, and split moving object for decrease searching space an Estimated-Split algorithm that minimizes the volume of MBRs is designed and simulated. Our experimental evaluation confirms the effectiveness and efficiency of our proposed splitting policy.