• Journal of Digital Convergence
• /
• v.11 no.2
• /
• pp.243-248
• /
• 2013

This paper proposes an effective method using Adaptive Boosting to track a person's head in complex background. By only one way to feature extraction methods are not sufficient for modeling a person's head. Therefore, the method proposed in this paper, several feature extraction methods for the accuracy of the detection head running at the same time. Feature Extraction for the imaging of the head was extracted using sub-region and Haar wavelet transform. Sub-region represents the local characteristics of the head, Haar wavelet transform can indicate the frequency characteristics of face. Therefore, if we use them to extract the features of face, effective modeling is possible. In the proposed method to track down the man's head from the input video in real time, we ues the results after learning Harr-wavelet characteristics of the three types using AdaBoosting algorithm. Originally the AdaBoosting algorithm, there is a very long learning time, if learning data was changes, and then it is need to be performed learning again. In order to overcome this shortcoming, in this research propose efficient method using cascade AdaBoosting. This method reduces the learning time for the imaging of the head, and can respond effectively to changes in the learning data. The proposed method generated classifier with excellent performance using less learning time and learning data. In addition, this method accurately detect and track head of person from a variety of head data in real-time video images.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1998.03a
• /
• pp.172-177
• /
• 1998

An experimental study on the prediction of grinding wheel wear by dual detection methods was conducted by the laser displacement and acoustic emission(AE) system. The laser displacement sensor was located above the head of the grinding wheel and the AE sensor was set under the workpiece, where the wheel were condition can be detected. It was found that the dual detection methods by laser displacement system and AE system made it possible to predict the wheel wear. From the experiments, the root mean square(RMS) values both methods was found to be proportional to the grinding wheel wear.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.987-990
• /
• 2005

An important issue in the field of face recognitions and man-machine interfaces is an automatic detection of faces in visual scenes. it should be computationally fast enough to allow an online detection. In this paper we describe our ongoing work on face detection that models the face appearance by edge orientation and color distribution. We show that edge orientation is a powerful feature to describe objects like faces. We present a method for face region detection using edge orientation and a method for face feature detection using local color information. We demonstrate the capability of our detection method on an image database of 1877 images taken from more than 700 people. The variations in head size, lighting and background are considerable, and all images are taken using low-end cameras. Experimental results show that the proposed scheme achieves 94% detection rate with a resonable amount of computation time.

• ETRI Journal
• /
• v.39 no.5
• /
• pp.652-661
• /
• 2017

A mention has a noun or noun phrase as its head and constructs a chunk that defines any meaning, including a modifier. Mention detection refers to the extraction of mentions from a document. In mentions, coreference resolution refers to determining any mentions that have the same meaning. Pointer networks, which are models based on a recurrent neural network encoder-decoder, outputs a list of elements corresponding to an input sequence. In this paper, we propose mention detection using pointer networks. This approach can solve the problem of overlapped mention detection, which cannot be solved by a sequence labeling approach. The experimental results show that the performance of the proposed mention detection approach is F1 of 80.75%, which is 8% higher than rule-based mention detection, and the performance of the coreference resolution has a CoNLL F1 of 56.67% (mention boundary), which is 7.68% higher than coreference resolution using rule-based mention detection.

• Korean Journal of Head & Neck Oncology
• /
• v.11 no.2
• /
• pp.173-177
• /
• 1995

Multiple primary malignant neoplasms occur relatively frequently today and are important especially in the head and neck area for they usually carry a bad prognosis. Detection of a synchronous primary tumor at the time of initial work-up is crucial both for management and final outcome. The first case was a T1 hypopharyngeal cancer with a mid-esophageal second primary who complained of a huge neck node. The second case was a T3 hypopharyngeal cancer who was initially seen by the chest surgeons for a large lower esophageal tumor. The third case was a patient previously operated for stomach adenocarcinoma three years ago, who had newly developed symptoms like dysphagia and hoarseness, and was diagnosed as hypopharyngeal T3 with oropharyngeal second primary cancer. Three cases were all heavy smokers and had histories of heavy alcohol consumption. They were all treated at the same sitting by en-block resection of the involved organs and postoperative radiation therapy. The authors have recently experienced 3 cases of synchronous second primary cancers in association with hypopharyngeal cancer and a report is made.

• Korean Journal of Head & Neck Oncology
• /
• v.11 no.2
• /
• pp.167-172
• /
• 1995

Pharyngocutaneous fistula(PCF) is one of the complications following total laryngectomy in laryngeal and hypopharyngeal cancer. Fistula lead to delayed wound healing, more serious complications such as carotid blow-out, prolonged hospitalization, significant patient morbidity and occasional mortality. Identification of patients at high risk for fistula formation, appropriate preventive measures, and appropriate management are the head and neck surgeon's responsibility. So we analyzed the clinical data of pharyngocutaneous fistula which was developed after total laryngectomy. Following results were obtained: 1) Occurrence of PCF increases with salvage surgery compared to curative surgery. 2) Early detection and effective management of PCF are the key factors to decreasing the hospitalization period. 3) Constructing a pharyngostoma seems to be an ideal method of preventing dangerous complications and ultimately closing the fistula. 4) Simultaneous reconstruction is necessary in the high risk group.

• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.345-349
• /
• 2023

In this paper, we propose a study on the development of deep learning structure for defective pixel detection of next-generation smart LED display board using imaging device. In this research, a technique utilizing imaging devices and deep learning is introduced to automatically detect defects in outdoor LED billboards. Through this approach, the effective management of LED billboards and the resolution of various errors and issues are aimed. The research process consists of three stages. Firstly, the planarized image data of the billboard is processed through calibration to completely remove the background and undergo necessary preprocessing to generate a training dataset. Secondly, the generated dataset is employed to train an object recognition network. This network is composed of a Backbone and a Head. The Backbone employs CSP-Darknet to extract feature maps, while the Head utilizes extracted feature maps as the basis for object detection. Throughout this process, the network is adjusted to align the Confidence score and Intersection over Union (IoU) error, sustaining continuous learning. In the third stage, the created model is employed to automatically detect defective pixels on actual outdoor LED billboards. The proposed method, applied in this paper, yielded results from accredited measurement experiments that achieved 100% detection of defective pixels on real LED billboards. This confirms the improved efficiency in managing and maintaining LED billboards. Such research findings are anticipated to bring about a revolutionary advancement in the management of LED billboards.

• The Journal of Korea Robotics Society
• /
• v.5 no.2
• /
• pp.152-159
• /
• 2010

Finding a head of a person in a scene is very important for taking a well composed picture by a robot photographer because it depends on the position of the head. So in this paper, we propose a robust head tracking algorithm using a hybrid of an omega shape tracker and local binary pattern (LBP) AdaBoost face detector for the robot photographer to take a fine picture automatically. Face detection algorithms have good performance in terms of finding frontal faces, but it is not the same for rotated faces. In addition, when the face is occluded by a hat or hands, it has a hard time finding the face. In order to solve this problem, the omega shape tracker based on active shape model (ASM) is presented. The omega shape tracker is robust to occlusion and illuminationchange. However, whenthe environment is dynamic,such as when people move fast and when there is a complex background, its performance is unsatisfactory. Therefore, a method combining the face detection algorithm and the omega shape tracker by probabilistic method using histograms of oriented gradient (HOG) descriptor is proposed in this paper, in order to robustly find human head. A robot photographer was also implemented to abide by the 'rule of thirds' and to take photos when people smile.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.3
• /
• pp.265-272
• /
• 2009

In this paper, a simple optical method that uses an infrared(IR) cut filter is proposed to minimize variation of eye image by external infrared(IR) sources in a video based head mounted eye tracking system that is used in the field of sports. For this, the IR cut filter is attached to a head mount of the eye tracking system, and the camera with an IR LED is located between the IR cut filter and eye. In this structure, external IR is blocked by the IR cut filter, and the IR intensity on the eye can be controlled by the IR LED. Therefore, the illumination condition of the camera to capture the eye can be stable without being affected by external IR illuminations. To verify the proposed idea, variation of the eye image and intensity of the IR with/without the IR cut filter is measured under various illumination conditions. The measured data show that the IR cut filter method can block external IR effectively, and complex pupil detection algorithms can be replaced by a simple binarized method.

• Imaging Science in Dentistry
• /
• v.32 no.2
• /
• pp.99-105
• /
• 2002

Purpose: To evaluate the diagnostic accuracy of transcranial radiographs and magnetic resonance imaging (MRI) of the temporomandibular joint (TMJ) in the assessment of osseous changes of the condylar head and articular eminence. Materials and Methods: Osseous changes of the TMJ were evaluated in forty-three patients. Osseous changes of the condylar head and articular eminence were observed in 41 joints and 64 joints, respectively on transcranial radiographs, and 48 joints and 59 joints, respectively on MRI. Results: The flattening, sclerosis, erosion, and osteophyte formation of the condylar heads were observed in 36.6%, 43.9%, 12.2%, and 7.3%, respectively on transcranial radiographs compared with 35.4%, 20.8%, 37.5%, and 6.3%, respectively on MRI. While, the flattening, sclerosis, and erosion of the articular eminences were observed in 26.6%, 67.2%, and 6.2%, respectively on transcranial radiographs compared with 32.2%, 59.3%, and 8.5%, respectively on MRI. Conclusion: There were no statistical differences between transcranial radiographs and MRI scans in the detection of osseous changes of the TMJ. However, MRI scans were superior to the transcranial radiographs in the detection of erosion of the condylar head (p<0.01).