• Journal of Broadcast Engineering
• /
• v.21 no.6
• /
• pp.913-919
• /
• 2016

In this paper, we propose the visual tracking algorithm that takes advantage of multiple random walkers. We first show the tracking method based on support vector machine as [1] and suggest a method that suppresses feature vectors extracted from backgrounds while preserve features vectors from foregrounds. We also show how to discriminate between foregrounds and backgrounds. Learned by reducing influences of backgrounds, support vector machine can clearly distinguish foregrounds and backgrounds from the image whose target objects are similar to backgrounds and occluded by another object. Thus, the algorithm can track target objects well. Furthermore, we introduce a simple method improving tracking speed. Finally, experiments validate that proposed algorithm yield better performance than the state-of-the-art trackers on the widely-used benchmark dataset with high speed.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.4 s.36
• /
• pp.113-121
• /
• 2005

Face detection in real-time video constitutes one of the major trend in face recognition. In this paper, we propose a face detection algorithm using the skin color and Haar-like feature in real-time video. The proposed algorithm is followed by three sequences; First, moving objects are detected by difference-method in YCbCr coordinates, and then by using Haar-like features, face candidate regions of the moving objects is selected. Finally we extract the most possible face candidates by comparing the pixel values of face candidates with the skin color. In order to prevent a mistake. we use similar features or skin color to detect a face by selecting a adaptive ROI and improve the processing speed in real-time video. The computer simulation shows the validity of the proposed method that the processing speed is improved by 30% than previous works and the detection success rate is 96.8%.

• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.135-142
• /
• 2023

In addressing the prominent issues of climate change, resource scarcity, and environmental pollution associated with household waste, extensive research has been conducted on intelligent waste classification methods. These efforts range from traditional classification algorithms to machine learning and neural networks. However, challenges persist in effectively classifying waste in diverse environments and conditions due to insufficient datasets, increased complexity in neural network architectures, and performance limitations for real-world applications. Therefore, this paper proposes i-YOLOX as a solution for rapid classification and improved accuracy. The proposed model is evaluated based on network parameters, detection speed, and accuracy. To achieve this, a dataset comprising 10,000 samples of household waste, spanning 17 waste categories, is created. The i-YOLOX architecture is constructed by introducing the Involution channel convolution operator and the Convolution Branch Attention Module (CBAM) into the YOLOX structure. A comparative analysis is conducted with the performance of the existing YOLO architecture. Experimental results demonstrate that i-YOLOX enhances the detection speed and accuracy of waste objects in complex scenes compared to conventional neural networks. This confirms the effectiveness of the proposed i-YOLOX architecture in the detection and classification of multiple household waste objects.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2019.05a
• /
• pp.18-20
• /
• 2019

In this paper, in order to prevent accidents when deviating from the lane during high-speed self-driving, we are going to design a warning system that will sound an alarm after recognizing the surrounding situation with a $360^{\circ}$ camera. Accidents often occur while driving on self-driving cars because they try to change lanes excessively or fail to recognize people, animals and objects that appear suddenly when driving at high speeds. The government wants to identify the surrounding situation with cameras when driving off a lane during high-speed autonomous driving, and to create a car that sounds a warning system through a lane departure sensor on the underside of the vehicle to reduce various accidents that occur during self-driving and to have a safer driving system.

• Journal of Korea Multimedia Society
• /
• v.5 no.6
• /
• pp.625-637
• /
• 2002

The Internet had been taken properties which were both a slow speed and low-bandwidth of the first time. It was impossible to transmit large multimedia data because of a Internet properties. After tens of Year Passed, many people began to transmit multimedia data in the Internet which had fast-speed and high-bandwidth. Some people want to team classes to be taught in school through the Internet. And, educators wish to use multimedia data in the educational field. Due to previous reasons, many parts in the educational field had tried to join real-time objects to multimedia objects. But, real-time data didn’t appear easily due to many faults like network errors and system bugs if we want to control data in order to teach and learn classes. Then, this paper separates real-time and unreal-time data and controls these data that aye made a time scheduler to have time information in the extended SMIL-file for deceasing these faults.

• Journal of the Korean Society for Marine Environment & Energy
• /
• v.4 no.1
• /
• pp.47-62
• /
• 2001

An experimental study was peformed to provide the numerical models to predict the physical fate of marine debris with the key information concerning their trifling and sinkage on the sea. For hygroscopic materials their floating times in a pseudo ocean environment were measured, and for non-hygroscopic objects their drifting limes in a certain wave condition were measured. For sinking materials or objects became to be sinkable free falling experiment was performed to evaluate their falling speed. Summarized information was condensed into a numerical model to predict the most possible region of a significant marine debris accumulation, and the seas connected to the Han river and the Nak-Dong river were considered for the simulation.

• The Mathematical Education
• /
• v.57 no.3
• /
• pp.311-328
• /
• 2018

In school mathematics, the concept of an average is not a concept that is limited to a unit of statistics. In particular, high school students will learn about arithmetic mean and geometric mean in the process of learning absolute inequality. In calculus learning, the concept of average is involved when learning the concept of average speed. The arithmetic mean is the same as the procedure used when students mean the test scores. However, the procedure for obtaining the geometric mean differs from the procedure for the arithmetic mean. In addition, if the arithmetic mean and the geometric mean are the discrete quantity, then the mean rate of change or the average speed is different in that it considers continuous quantities. The average concept that students learn in school mathematics differs in the quantitative nature of procedures and objects. Nevertheless, it is not uncommon to find out how students construct various mathematical concepts into mathematical knowledge. This study focuses on this point and conducted the interviews of the students(three) in the second grade of high school. And the expression of students in the process of average concept formation in arithmetic mean, geometric mean, average speed. This study can be meaningful because it suggests practical examples to students about the assertion that various scholars should experience various properties possessed by the average. It is also meaningful that students are able to think about how to construct the mean conceptual properties inherent in terms such as geometric mean and mean speed in arithmetic mean concept through interview data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.8
• /
• pp.656-663
• /
• 2001

This paper presents a sensor fusion method to recognize a cylindrical object a CCD camera, a laser slit beam and ultrasonic sensors on a pan/tilt device. For object recognition with a vision sensor, an active light source projects a stripe pattern of light on the object surface. The 2D image data are transformed into 3D data using the geometry between the camera and the laser slit beam. The ultrasonic sensor uses an ultrasonic transducer array mounted in horizontal direction on the pan/tilt device. The time of flight is estimated by finding the maximum correlation between the received ultrasonic pulse and a set of stored templates - also called a matched filter. The distance of flight is calculated by simply multiplying the time of flight by the speed of sound and the maximum amplitude of the filtered signal is used to determine the face angle to the object. To determine the position and the radius of cylindrical objects, we use a statistical sensor fusion. Experimental results show that the fused data increase the reliability for the object recognition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.11
• /
• pp.5436-5458
• /
• 2017

Automatic Video Surveillance System (AVSS) has become important to computer vision researchers as crime has increased in the twenty-first century. As a new branch of AVSS, baggage detection has a wide area of security applications. Some of them are, detecting baggage in baggage restricted super shop, detecting unclaimed baggage in public space etc. However, in this paper, a detection & classification framework of baggage is proposed. Initially, background subtraction is performed instead of sliding window approach to speed up the system and HSI model is used to deal with different illumination conditions. Then, a model is introduced to overcome shadow effect. Then, occlusion of objects is detected using proposed mirroring algorithm to track individual objects. Extraction of rotational signal descriptor (SP-RSD-HOG) with support plane from Region of Interest (ROI) add rotation invariance nature in HOG. Finally, dynamic human body parameter setting approach enables the system to detect & classify single or multiple pieces of carried baggage even if some portions of human are absent. In baggage detection, a strong classifier is generated by boosting similarity measure based multi layer Support Vector Machine (SVM)s into HOG based SVM. This boosting technique has been used to deal with various texture patterns of baggage. Experimental results have discovered the system satisfactorily accurate and faster comparative to other alternatives.

• Journal of Information Processing Systems
• /
• v.16 no.2
• /
• pp.261-276
• /
• 2020

Digital contents services are one of the topics that have been intensively studied in the media industry, where various semantic and ontology techniques are applied. However, query execution for ontology data is still inefficient, lack of sufficient extensible definitions for node relationships, and there is no specific semantic method fit for media data representation. In order to make the machine understand digital contents (DCs) data well, we analyze DCs data, including static data and dynamic data, and use ontology to specify and classify objects and the events of the particular objects. Then the formal representation method is proposed which not only redefines DCs data based on the technology of OWL/RDF, but is also combined with media segmentation methods. At the same time, to speed up the access mechanism of DCs data stored under the persistent database, an ontology-based DCs query solution is proposed, which uses the specified distance vector associated to a surveillance of semantic label (annotation) to detect and track a moving or static object.