• 방송공학회논문지
• /
• 제24권2호
• /
• pp.281-291
• /
• 2019

3D 카메라 기술 중에서 초점의 흐려짐을 이용한 깊이 추정은 카메라의 초점거리 평면의 물체는 선명한 상이 맺히지만 카메라의 초점거리 평면으로부터 멀어진 물체는 흐린 영상을 만들어낸다는 현상을 이용해 3D 깊이를 추정한다. 본 논문에서는 단일 카메라를 이용하여 촬영한 영상의 흐림 정도를 분석하여 3D 깊이를 추정하는 알고리즘을 연구하였다. 단일 카메라의 1 개의 영상 또는 단일 카메라의 초점이 서로 다른 2 개의 영상을 사용하여 초점의 흐려짐을 이용한 3D 깊이를 추정하는 방법을 통해 최적화된 피사체 범위를 도출하였다. 1 개의 영상을 이용한 깊이 추정에서는 스마트폰 카메라와 DSLR 카메라 모두 250 mm의 초점거리를 사용하는 것이 가장 좋은 성능을 보였다. 2개의 영상을 이용한 깊이 추정에서는 스마트폰 카메라 영상은 150 mm와 250 mm로 그리고 DSLR 카메라 영상은 200 mm와 300 mm로 초점거리를 설정하였을 때 가장 좋은 3D 깊이 추정 유효 범위를 갖는 것으로 나타났다.

• 대한의용생체공학회:의공학회지
• /
• 제31권5호
• /
• pp.407-411
• /
• 2010

During the cardiopulmonary resuscitation (CPR), the correct chest compression depth and period are very important to increase the resuscitation possibility. For the feedback of chest compression depth, the depth monitoring device based on the accelerometer is developed and widely used. But this method tends to overestimate the compression depth on the bed. To overcome this limitation, the chest compression depth estimation method using two accelerometers is suggested With the additional accelerometer between the patient and mattress on the bed, the compression of the mattress is also measured and it is used to compensate the overestimation error. The experimental results show that the single accelerometer estimates as 61.4mm for the actual compression depth of 43.6mm on the mattress. The depth estimation with the dual accelerometer was 44.6mm which is close to the actual depth. With the automatic zeroing in every single compression, the integration error for the depth can be reduced. The dual accelerometer method is effective to increase the accuracy of the chest compression depth estimation.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 신호처리소사이어티 추계학술대회 논문집
• /
• pp.171-174
• /
• 2003

In this paper, we propose the method of stereo images composition using adaptive dense disparity estimation. For the correct composition of stereo image and 3D virtual object, we need correct marker position and depth information. The existing algorithms use position information of markers in stereo images for calculating depth of calibration object. But this depth information may be wrong in case of inaccurate marker tracking. Moreover in occlusion region, we can't know depth of 3D object, so we can't composite stereo images and 3D virtual object. In these reasons, the proposed algorithm uses adaptive dense disparity estimation for calculation of depth. The adaptive dense disparity estimation is the algorithm that use pixel-based disparity estimation and the search range is limited around calibration object.

• 한국정밀공학회지
• /
• 제12권11호
• /
• pp.52-62
• /
• 1995

In the laser surface hardening process the geometrical parameters, especially the depth, of the hardened layer are utilized to assess the integrity of the hardening layer quality. Monitoring of this geometrical parameter ofr on-line process control as well as for on-line quality evaluation, however, is an extremely difficult problem because the hardening layer is formed beneath a material surface. Moreover, the uncertainties in monitoring the depth can be raised by the inevitable use of a surface coating to enhance the processing efficiency and the insufficient knowledge on the effects of coating materials and its thicknesses. The paper describes the extimation results using neural network to estimate the hardening layer depth from measured surface temperanture and process variables (laser beam power and feeding velocity) under various situations. To evaluate the effec- tiveness of the measured temperature in estimating the harding layer depth, estimation was performed with or without temperature informations. Also to investigate the effects of coating thickness variations in the real industry situations, in which the coating thickness cannot be controlled uniform with good precision, estimation was done over only uniformly coated specimen or various thickness-coated specimens. A series of hardening experiments were performed to find the relationships between the hardening layer depth, temperature and process variables. The estimation results show the temperature informations greatly improve the estimation accuracy over various thickness-coated specimens.

• Structural Monitoring and Maintenance
• /
• 제9권4호
• /
• pp.373-388
• /
• 2022

A USW based diagnostic procedure is presented for estimating the depth of surface-breaking cracks. The diagnosis is demonstrated on seven lab-scale SFRC beam specimens, which are subjected to the CMOD controlled three-point bending test to create real bending cracks. Then, the recorded multiple ultrasonic signals are examined with the signal processing techniques, including wavelet transform and two-dimensional Fourier transform, to investigate the relationships between the crack depth and two diagnostic indices, namely the attenuation coefficient and dispersion index (DI). Finally, the reliabilities of these indices for depth estimation are verified with the visually measured crack depths as well as the crack features obtained with a digital image processing algorithm. It is found that the DI outperforms the attenuation coefficient in depth estimation, where this index displays good agreement with the visual inspection for 86% of the inspected specimens.

• 한국컴퓨터그래픽스학회논문지
• /
• 제28권3호
• /
• pp.45-54
• /
• 2022

영상의 깊이 추정은 다양한 영상 분석의 기반이 되는 기술이다. 딥러닝 모델을 활용한 분석 방법이 대두되면서, 영상의 깊이 추정 분야 또한 딥러닝을 활용하는 연구가 활발하게 이루어지고 있다. 현재 대부분의 딥러닝 영상 깊이 추정 모델들은 깨끗하고 이상적인 환경에서 학습되고 있다. 하지만 연무, 안개가 낀 열악한 환경에서도 깊이 추정 기술이 잘 동작할 수 있으려면 이러한 환경의 데이터를 포함하여야 한다. 하지만 열악한 환경의 영상을 충분히 확보하는 것이 어려운 실정이며, 불균일한 안개 데이터를 얻는 것은 특히 어려운 문제이다. 이를 해결하기 위해, 본 연구에서는 불균일 안개 영상 합성 방법과 이를 활용한 단안 기반의 깊이 추정 딥러닝 모델의 학습을 제안한다. 안개가 주로 실외에서 발생하는 것을 고려하여, 실외 위주의 데이터 세트를 구축한다. 그리고 실험을 통해 제안된 방법으로 학습된 모델이 합성 데이터와 실제 데이터에서 깊이를 잘 추정하는 것을 보인다.

• 디지털산업정보학회논문지
• /
• 제4권2호
• /
• pp.15-19
• /
• 2008

In this paper, we propose a new algorithm of image depth detection using motion estimation and object tracking. In industry, robots are used for automobile, conveyer system, etc. But, these have much necessary time. Thus, in this paper, we develop the efficient method of image depth detection based on motion estimation and object tracking.

• ETRI Journal
• /
• 제44권1호
• /
• pp.38-50
• /
• 2022

This paper presents a study on the use of encoder-derived features in decoder-side depth estimation. The scheme of multiview video encoding does not require the transmission of depth maps (which carry the geometry of a three-dimensional scene) as only a set of input views and their parameters are compressed and packed into the bitstream, with a set of features that could make it easier to estimate geometry in the decoder. The paper proposes novel recursive block splitting for the feature extraction process and evaluates different scenarios of feature-driven decoder-side depth estimation, performed by assessing their influence on the bitrate of metadata, quality of the reconstructed video, and time of depth estimation. As efficient encoding of multiview sequences became one of the main scopes of the video encoding community, the experimental results are based on the "geometry absent" profile from the incoming MPEG Immersive video standard. The results show that the quality of synthesized views using the proposed recursive block splitting outperforms that of the state-of-the-art approach.

• 한국멀티미디어학회논문지
• /
• 제20권11호
• /
• pp.1711-1719
• /
• 2017

In this paper, we propose a method of the zoom motion estimation for the depth video coding. The proposed method calculates the zoom ratio using the average of the depth values in the current block and in the reference block. It resizes the reference block by the zoom ratio and interpolates the reference block to size of the current block. It compares the current block with the reference block that is obtained by subtracting the average of pixels from the current block to the reference block in order to find the reference block that is the best closest one to the current block. The results of the simulation for the proposed method show that the motion estimation errors are significantly reduced.

• 한국멀티미디어학회논문지
• /
• 제16권2호
• /
• pp.131-137
• /
• 2013

동영상의 줌 움직임 추정은 구현이 아주 복잡하다. 본 논문에서는 줌 움직임 추정을 구현하기 위하여 깊이 카메라와 색상 카메라를 동시에 이용하는 방법을 제안한다. 깊이 카메라로부터 현재블록과 참조블록 사이의 거리 정보를 얻고, 이 거리 정보로부터 두 블록사이의 줌 비율을 계산한다. 줌 비율에 맞게 참조블록을 확대 또는 축소시켜 줌으로서 움직임 추정 차신호를 줄일 수 있다. 따라서, 제안된 방법은 줌 움직임 추정을 위한 복잡도가 크지 않으면서 움직임 추정 정확도를 높이는 것이 가능하다. 모의실험을 바탕으로 제안된 방법의 움직임 추정 정확도를 측정하였으며, 기존 블록정합 방법에 비하여 움직임 추정 오차값이 크게 감소함을 확인하였다.