• International Journal of Advanced Culture Technology
• /
• 제10권2호
• /
• pp.252-259
• /
• 2022

Artificial intelligence is used in fusion with image processing techniques using cameras. Image processing technology is a technology that processes objects in an image received from a camera in real time, and is used in various fields such as security monitoring and medical image analysis. If such image processing reduces the accuracy of recognition, providing incorrect information to medical image analysis, security monitoring, etc. may cause serious problems. Therefore, this paper uses a mixture of YOLOv4-tiny model and image processing algorithm and uses the COCO dataset for learning. The image processing algorithm performs five image processing methods such as normalization, Gaussian distribution, Otsu algorithm, equalization, and gradient operation. For RGB images, three image processing methods are performed: equalization, Gaussian blur, and gamma correction proceed. Among the nine algorithms applied in this paper, the Equalization and Gaussian Blur model showed the highest object detection accuracy of 96%, and the gamma correction (RGB environment) model showed the highest object detection rate of 89% outdoors (daytime). The image binarization model showed the highest object detection rate at 89% outdoors (night).

• 한국멀티미디어학회논문지
• /
• 제23권4호
• /
• pp.525-538
• /
• 2020

A practical animal face classification system that classifies animals in image and video data is considered as a pivotal topic in machine learning. In this research, we are proposing a novel method of fully connected dual Deep Convolutional Neural Network (DCNN), which extracts and analyzes image features on a large scale. With the inclusion of the state of the art Batch Normalization layer and Exponential Linear Unit (ELU) layer, our proposed DCNN has gained the capability of analyzing a large amount of dataset as well as extracting more features than before. For this research, we have built our dataset containing ten thousand animal faces of ten animal classes and a dual DCNN. The significance of our network is that it has four sets of convolutional functions that work laterally with each other. We used a relatively small amount of batch size and a large number of iteration to mitigate overfitting during the training session. We have also used image augmentation to vary the shapes of the training images for the better learning process. The results demonstrate that, with an accuracy rate of 92.0%, the proposed DCNN outruns its counterparts while causing less computing costs.

• 한국융합학회논문지
• /
• 제10권9호
• /
• pp.323-332
• /
• 2019

본 연구는 전겸구 등(2000)이 개발한 개정판 다차원적 대처척도(Revised Multidimensional Coping Scale; MCS-R)의 타당도와 신뢰도를 평가하기 위하여 실시된 방법론적 연구이다. 자료 수집은 금연집단상담에 참여한 대학생 198명을 대상으로 실시되었고, 탐색적 요인분석을 통해 구성타당도, 우울과 극복력 도구를 이용하여 수렴타당도를 확인하였다. MCS-R의 요인구조를 주성분 분석방법, 직교 회전으로 분석한 결과, 동일하게 13개의 요인이 확인되었으나, 정서적 진정, 긍정적 해석, 적극적 망각 3개의 요인에서 요인을 구성하는 문항이 다르게 나타났다. 그러나 13개의 요인 모두 요인적재량이 0.40이상으로 나타났고, 13의 요인에 의해 설명된 총 분산은 69.7%로 높았고, 각 하위영역의 내적 일관성(0.66-0.94)과 검사-재검사 신뢰도(0.44-0.85)는 적절하였다. 개정판 다차원적 대처척도는 개인의 대처방식을 측정하기에 적합한 도구이며, 향후 스트레스 대처를 위한 효과적 개입을 위한 계획 수립에 기여할 것으로 기대된다.

• 대한두개안면성형외과학회지
• /
• 제25권1호
• /
• pp.17-21
• /
• 2024

Background: The pain caused by local anesthetic injection can lead to patient anxiety prior to surgery, potentially necessitating sedation or general anesthesia during the excision procedure. In this study, we aim to compare the pain relief efficacy and safety of using a digital automatic anesthetic injector for local anesthesia. Methods: Thirty-three patients undergoing excision of a benign soft tissue tumor under local anesthesia were prospectively enrolled from September 2021 to February 2022. A single-blind, randomized controlled study was conducted. Patients were divided into two groups by randomization: the experimental group with digital automatic anesthetic injector method (I-JECT group) and the control group with conventional injection method. Before surgery, the Amsterdam preoperative anxiety information scale was used to measure the patients' anxiety. After local anesthetic was administered, the Numeric Pain Rating Scale was used to measure the pain. The amount of anesthetic used was divided by the surface area of the lesion was recorded. Results: Seventeen were assigned to the conventional group and 16 to the I-JECT group. The mean Numeric Pain Rating Scale was 1.75 in the I-JECT group and 3.82 in conventional group. The injection pain was lower in the I-JECT group (p< 0.01). The mean Amsterdam preoperative anxiety information scale was 11.00 in the I-JECT group and 9.65 in conventional group. Patient's anxiety did not correlate to injection pain regardless of the method of injection (p= 0.47). The amount of local anesthetic used per 1 cm² of tumor surface area was 0.74 mL/cm² in the I-JECT group and 2.31 mL/cm² in the conventional group. The normalization amount of local anesthetic was less in the I-JECT group (p< 0.01). There was no difference in the incidence of complications. Conclusion: The use of a digital automatic anesthetic injector has shown to reduce pain and the amount of local anesthetics without complication.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권2호
• /
• pp.763-774
• /
• 2015

Hand posture recognition has been a wide region of applications in Human Computer Interaction and Computer Vision for many years. The problem arises mainly due to the high dexterity of hand and self-occlusions created in the limited view of the camera or illumination variations. To remedy these problems, a hand posture recognition method using 3-D point cloud is proposed to explicitly utilize 3-D information from depth maps in this paper. Firstly, hand region is segmented by a set of depth threshold. Next, hand image normalization will be performed to ensure that the extracted feature descriptors are scale and rotation invariant. By robustly coding and pooling 3-D facets, the proposed descriptor can effectively represent the various hand postures. After that, SVM with Gaussian kernel function is used to address the issue of posture recognition. Experimental results based on posture dataset captured by Kinect sensor (from 1 to 10) demonstrate the effectiveness of the proposed approach and the average recognition rate of our method is over 96%.

• Journal of Advanced Marine Engineering and Technology
• /
• 제27권7호
• /
• pp.852-861
• /
• 2003

In this paper, in order to improve the disadvantages of the fixed design-parameter fuzzy PID controller. a new fuzzy PID controller named a variable design-parameter fuzzy PID controller is suggested. The main characteristic of the suggested controller is to adjust design-parameters of the controller by comparing magnitudes between fuzzy controller inputs at each sampling time when controller inputs are measured. As a result. all fuzzy input partitioned spaces converge within a time-varying normalization scale. and the resultant PID control action can always be applied precisely regardless of operating input magnitudes. In order to verify the effectiveness of the suggested controller. several a computer simulations for a nonlinear system are executed and the control parameters of the variable design-parameter fuzzy PID controller are throughly analyzed.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -2
• /
• pp.869-872
• /
• 2000

Although many content-based image retrieval systems using shape feature have tried to cover rotation-, position- and scale-invariance between images, there have been problems to cover three kinds of variance at the same time. In this paper, we introduce new approach to extract shape feature from image using MBR(Minimum Bounding Rectangle). The proposed method scans image for extracting MBR information and, based on MBR information, compute contour information that consists of 16 points. The extracted information is converted to specific values by normalization and rotation. The proposed method can cover three kinds of invariance at the same time. We implemented our method and carried out experiments. We constructed R*_tree indexing structure, perform k-nearest neighbor search from query image, and demonstrate the capability and usefulness of our method.

• 한국재난정보학회:학술대회논문집
• /
• 한국재난정보학회 2022년 정기학술대회 논문집
• /
• pp.107-108
• /
• 2022

본 연구에서는 Arc GIS의 Network Analysis로 119안전센터의 재난 대응 권역을 설정해 재난 취약 면적을 계산하고 지역자율방재단의 재난피해 완충효과를 분석했다. 모든 값은 Min-Max Normalization 되어 동일한 Scale로 계산되었다. 지역자율방재단은 재난피해 완충 대책으로써 유의미한 효과가 있음을 확인했다. 지속적인 지역자율방재단의 활성화는 주민 참여, 지역특화적 재난 방재 대책 수립에 효과적이다.

• 전자공학회논문지CI
• /
• 제45권1호
• /
• pp.25-36
• /
• 2008

눈 검출은 눈 동공의 정 중앙의 위치를 찾아내는 작업을 의미하며, 얼굴 인식 및 관련된 응용 분야 등에서 필요한 작업이다. 현재까지 보고된 대부분의 눈 검출 방법의 경우 성공적인 적용을 위해서는 여전히 정확도 및 검출 속도의 개선을 필요로 한다. 본 논문에서는 큰 계산량의 부담이 없는 다중 해상도 가버 특징 벡터를 이용한 강인한 눈 검출 방법을 제안한다. 가버 특징 벡터를 사용한 눈 검출은 EBGM 등에서 이미 이용되고 있다. 그런데, RBGM 등에서 사용한 눈 검출 방법은 초기값에 민감하고 조명, 자세 등에 강인하지 못하여, 만족할 만한 검출률을 얻기 위해서는 광범위한 탐색 범위가 필요하다. 이는 계산량의 상당한 증가를 초래한다. 본 논문에서 제안한 눈 검출 방법은 다중 해상도 접근 방법을 활용한다. 먼저, 원래 해상도 얼굴 이미지를 다운샘플링하여 얻은 저해상도 얼굴 이미지에서, 초기 추정 눈 위치에서의 가버 특징 벡터와 해당 해상도의 눈에 대한 가버 특징 벡터 모델과의 가버젯 유사도를 이용하여 눈 위치를 검출한다. 이후 검출된 눈 위치를 업스케일링하여 상위 해상도의 얼굴 이미지에서의 눈 위치 초기값으로 취하고 앞 단계에서처럼 가버젯 유사도를 이용하여 눈을 검출한다. 이 과정을 반복하여 최종적으로 원래 해상도 얼굴 이미지에서의 눈 위치를 확정한다. 또한, 본 논문에서는 제안된 다중 해상도 접근 방법이 조명에 대해서도 보다 강인하도록 하는 데 효과적인 조명 정규화 기법을 제안하고, 이를 다중 해상도 접근 방법의 전처리 단계에 추가적으로 적용함으로써 눈 검출 성공률을 더욱 개선하였다. 실험을 통해, 본 논문에서 제안한 다중스케일 가버 특징 벡터 기반 눈 검출 방법은 계산량을 크게 증가 시키지 않으면서 기존 연구들에서 보고된 다른 눈 검출 방법에 비해 정확도가 개선된 검출 방법이며, 자세 및 조명 변화에 대해서도 강인하다는 것을 확인하였다.

• ETRI Journal
• /
• 제37권2호
• /
• pp.395-405
• /
• 2015

We address the problem of uncooperative person recognition through continuous monitoring. Multiple modalities, such as face, height, clothes color, and voice, can be used when attempting to recognize a person. In general, not all modalities are available for a given frame; furthermore, only some modalities will be useful as some frames in a video sequence are of a quality that is too low to be able to recognize a person. We propose a method that makes use of stochastic information updates of temporal modalities and environment estimators to improve person recognition performance. The environment estimators provide information on whether a given modality is reliable enough to be used in a particular instance; such indicators mean that we can easily identify and eliminate meaningless data, thus increasing the overall efficiency of the method. Our proposed method was tested using movie clips acquired under an unconstrained environment that included a wide variation of scale and rotation; illumination changes; uncontrolled distances from a camera to users (varying from 0.5 m to 5 m); and natural views of the human body with various types of noise. In this real and challenging scenario, our proposed method resulted in an outstanding performance.