• Journal of the Korean Society for Precision Engineering
• /
• v.25 no.4
• /
• pp.39-46
• /
• 2008

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.6
• /
• pp.718-723
• /
• 2011

In this paper, we propose a visual multi-touch air input device using vision cameras. The implemented device provides a barehanded interface which copes with the multi-touch operation. The proposed device is easy to apply to the real-time systems because of its low computational load and is cheaper than the existing methods using glove data or 3-dimensional data because any additional equipment is not required. To do this, first, we propose an image processing algorithm based on the HSV color model and the labeling from obtained images. Also, to improve the accuracy of the recognition of hand gestures, we propose a motion recognition algorithm based on the geometric feature points, the skeleton model, and the Kalman filter. Finally, the experiments show that the proposed device is applicable to remote controllers for video games, smart TVs and any computer applications.

• Journal of Digital Convergence
• /
• v.12 no.2
• /
• pp.321-326
• /
• 2014

The needs for IT accessibility for disabled persons has increased for recent years. So, it is very important to support multi-modal interfaces, such as voice and vision recognition, TTS, etc. for disabled persons. In this paper, we deal with IT accessibility issues of home networks and show our implemented home network control system model with multi-modal interfaces including voice recognition and animated user interfaces.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.45-59
• /
• 2023

Weakly supervised object localization (WSOL) is a task of localizing an object in an image using only image-level labels. Previous studies have followed the conventional class activation mapping (CAM) pipeline. However, we reveal the current CAM approach suffers from problems which cause original CAM could not capture the complete defects features. This work utilizes a convolutional neural network (CNN) pretrained on image-level labels to generate class activation maps in a multi-scale manner to highlight discriminative regions. Additionally, a vision transformer (ViT) pretrained was treated to produce multi-head attention maps as an auxiliary detector. By integrating the CNN-based CAMs and attention maps, our approach localizes defective regions without requiring bounding box or pixel-level supervision during training. We evaluate our approach on a dataset of apple images with only image-level labels of defect categories. Experiments demonstrate our proposed method aligns with several Object Detection models performance, hold a promise for improving localization.

• The KIPS Transactions:PartB
• /
• v.17B no.6
• /
• pp.429-436
• /
• 2010

This paper presents a method to extract 3D outlines of objects in an image obtained from a monocular vision. After detecting the general outlines of the object by MOPS(Multi-Scale Oriented Patches) -algorithm and we obtain their spatial coordinates. Simultaneously, it obtains the space-coordinates with feature points to be immanent within the outlines of objects through SIFT(Scale Invariant Feature Transform)-algorithm. It grasps a form of objects to join the space-coordinates of outlines and SIFT feature points. The method which is proposed in this paper, it forms general outlines of objects, so that it enables a rapid calculation, and also it has the advantage capable of collecting a detailed data because it supplies the internal-data of outlines through SIFT feature points.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.15 no.5
• /
• pp.557-564
• /
• 2012

This paper presents UGV localization using multi-sensor fusion based on federated filter in outdoor environments. The conventional GPS/INS integrated system does not guarantee the robustness of localization because GPS is vulnerable to external disturbances. In many environments, however, vision system is very efficient because there are many features compared to the open space and these features can provide much information for UGV localization. Thus, this paper uses the scene matching and pose estimation based vision navigation, magnetic compass and odometer to cope with the GPS-denied environments. NR-mode federated filter is used for system safety. The experiment results with a predefined path demonstrate enhancement of the robustness and accuracy of localization in outdoor environments.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.317-319
• /
• 2021

In Korea, an intelligent transportation system(ITS) is established to efficiently operate traffic congestion on roads and is being used for traffic information collection and speed control systems. Currently, designated and dedicated lanes are in place to ensure traffic circulation and traffic safety, and systematic and accurate illegal vehicle crackdown systems with artificial intelligence technology are needed. In this study, we propose a vehicle number recognition model that can improve the efficiency of the traffic of designated vehicles. By applying computer vision technology, we are going to identify three-lane and four-lane multi-lane roads in real time and detect vehicle numbers by car to suggest ways to crack down on vehicles that violate the designated lane system.

• Smart Structures and Systems
• /
• v.15 no.6
• /
• pp.1583-1600
• /
• 2015

When natural disasters occur, including earthquakes, tsunamis, and debris flows, they are often accompanied by various types of damages such as the collapse of buildings, broken bridges and roads, and the destruction of natural scenery. Natural disaster detection and warning is an important issue which could help to reduce the incidence of serious damage to life and property as well as provide information for search and rescue afterwards. In this study, we propose a novel computer vision technique for debris flow detection which is feature-based that can be used to construct a debris flow event warning system. The landscape is composed of various elements, including trees, rocks, and buildings which are characterized by their features, shapes, positions, and colors. Unlike the traditional methods, our analysis relies on changes in the natural scenery which influence changes to the features. The "background module" and "monitoring module" procedures are designed and used to detect debris flows and construct an event warning system. The multi-criteria decision-making method used to construct an event warring system includes gradient information and the percentage of variation of the features. To prove the feasibility of the proposed method for detecting debris flows, some real cases of debris flows are analyzed. The natural environment is simulated and an event warning system is constructed to warn of debris flows. Debris flows are successfully detected using these two procedures, by analyzing the variation in the detected features and the matched feature. The feasibility of the event warning system is proven using the simulation method. Therefore, the feature based method is found to be useful for detecting debris flows and the event warning system is triggered when debris flows occur.

• Journal of Biosystems Engineering
• /
• v.31 no.5 s.118
• /
• pp.436-442
• /
• 2006

This research was conducted to design and develop a wired monitoring system for judging if sick or dead layers (SDL) exist in multi-tier layers battery (MLB) by machine vision, and to analyze its performance. In this study, 20 Brown Leghorn (Hi-Brown) layers aged 37 weeks old, were used as the experimental animals. The intensity of concern paid by layers on feed was over 90% during 5 minutes and 30 seconds after providing feed, and normal layers (NL) had been standing to take feed for that period. Therefore, in this study, the optimal judging time was set by this test result. The wired monitoring system developed was consisted of a driving device for carrying machine vision systems, a control program, a RS232 to RS485 convertor, an automatic positioning system, and an image capture system. An image processing algorithm was developed to find SDL in MLB by the processes of binary processing, erosion, expansion, labeling, and reckoning central coordinate of the captured images. The optimal velocity for driving unit was set up as 0.13 m/s by the test results for wired monitoring system, and the proximity switch was controlled not to be operated for 1.0 second after first image captured. The wired monitoring system developed was tested to evaluate the remote monitoring performance at lab-scale laying hen house. Results showed that its judgement success.ate on normal cage (without SDL) was 87% and that on abnormal cage (with SDL) was 90%, respectively. Therefore, it would be concluded that the wired monitoring system developed in this study was well suited to the purpose of this study.

• Journal of KIISE:Software and Applications
• /
• v.32 no.2
• /
• pp.85-98
• /
• 2005

Recently, a lot of researches about mobile vision using Personal Digital Assistant(PDA) has been attempted. Many CPUs for PDA are integer CPUs, which have no floating-computation component. It results in slow computation of the algorithms peformed by vision system or image processing, which have much floating-computation. In this paper, in order to resolve this weakness, we propose the Client(PDA)/server(PC) architecture which is connected to each other with a wireless LAN, and we construct the system with pipelining processing using two CPUs of the Client(PDA) and the Server(PC) in image sequence. The Client(PDA) extracts tentative text regions using Edge Density(ED). The Server(PC) uses both the Multi-1.aver Perceptron(MLP)-based texture classifier and Connected Component(CC)-based filtering for a definite text extraction based on the Client(PDA)'s tentativel99-y extracted results. The proposed method leads to not only efficient text extraction by using both the MLP and the CC, but also fast running time using Client(PDA)/server(PC) architecture with the pipelining processing.