• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1995년도 Proceedings of the Korea Automation Control Conference, 10th (KACC); Seoul, Korea; 23-25 Oct. 1995
• /
• pp.311-314
• /
• 1995

This paper presents a method for measuring tool wear parameters based on two dimensional image information. The tool wear images were obtained from an ITV camera with magnifying and lighting devices, and were analyzed using image processing techniques such as thresholding, noise filtering and boundary tracing. Thresholding was used to transform the captured gray scale image into a binary image for rapid sequential image processing. The threshold level was determined using a novel technique in which the brightness histograms of two concentric windows containing the tool wear image were compared. The use of noise filtering and boundary tracing to reduce the measuring errors was explored. Performance tests of the measurement precision and processing speed revealed that the direct method was highly effective in intermittent tool wear monitoring.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 추계종합학술대회 논문집
• /
• pp.1085-1088
• /
• 1999

This paper presents a method for soccer scene analysis and coordinate transformation from scene to ground model using a priori knowledge. First, the ground and spectator regions are separated, and various objects are extracted from the separated ground region. Second, an affine model is used for mapping the object positions on the soccer image into the position on the ground model. Problems regarding holes arising from mapping processing are solved using inverse mapping instead of a usual interpolation method. Experiments are performed on a PC using about 100 RGB images acquired at 240＊640 resolution and 3∼5 frames per second.

• 자원환경지질
• /
• 제32권5호
• /
• pp.495-502
• /
• 1999

Common-mid-point (CMP) seismic data on a dipping layer have have a stacking different from a horizontal layer velocity and the reflection points on data are dispersed to many positions. Therefore, the CMP data are not stacked well by the conventional stacking method using the horizontal layer velocity. The CMP gather can ideally stacked by applying dip-moveout(DMO) processing. Hence, modern seismic processing indludes DMO as an essential routine step. DMO processing techniques are broadly categorized by two, Fourier transform and integral methods, each of which has many different computational schemes. In this study, the dip-decomposition technique of the Fourier transform method is used to test the DMO effect on the synthetic scismic data generated for dipping structures. Each of constnat offset sections NMO corrected by using the layer velocity of the model and DMO processed. The resulting zero-offset sections for many offsets are stacked. The stacked sections with DMO processing show the structural boundaries of the models much better than those without DMO processing.

• Journal of Information Processing Systems
• /
• 제16권4호
• /
• pp.774-783
• /
• 2020

For the issue of subsurface target localization by reverse projection, a new approach of target localization with different distances based on symmetric sub-array multiple-input multiple-output (MIMO) radar is proposed in this paper. By utilizing the particularity of structure of the two symmetric sub-arrays, the received signals are jointly reconstructed to eliminate the distance information from the steering vectors. The distance-independent direction of arrival (DOA) estimates are acquired, and the localizations of subsurface targets with different distances are realized by reverse projection. According to the localization mechanism and application characteristics of the proposed algorithm, the grid zooming method based on spatial segmentation is used to optimize the locaiton efficiency. Simulation results demonstrate the effectiveness of the proposed localization method and optimization scheme.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.130.2-130
• /
• 2001

In this paper, we suggest an algorithm for the automatic measurement of concentricity and roundness using image processing technique. From the experimental results, we could find that the required measurement accuracy specification is sufficiently satisfied using our proposed method.

• 한국염색가공학회지
• /
• 제6권3호
• /
• pp.52-59
• /
• 1994

Polyester, polyester/cotton blend, nylon and cotton fabrics were ultraviolet cutting finished with padding method or exhaustion method using several UV absorbers. The transmittance of ultraviolet ray in the textiles can be greatly depressed by the processing and it is expected that human skin can be kept safe from sun-burn or damages by ultraviolet ray.

• 한국식품영양학회지
• /
• 제11권4호
• /
• pp.444-451
• /
• 1998

The changes of amine, nitrate and nitrite nitrogen in yellow corvenia were studied during its processing and storage in order to clarify the precursors of N-nitrosamine(NA) formation in the salted and dried yellow corvenia(Gulbi), prepared y using the different salting method like dry and brine salting by pure and curde salt. As a result, during the processing and storage of Gulbi, DMA and TMA contents were significantly increased in the yellow corvenia. And after 40 days storage the increase rate showed 25.7∼45.7, 3.3∼5.6 times higher than those of 0.3, 2.4mg/kg, respectively, while nitrite contents, during its processing and storage, were scarcely changed in the salted and dried yellow corvenia. During the processing and storage, of Gulbi, DMA and TMA contents were less produced in brine salted and dried yellow corvenia using crude salt than in sample prepared using were scarcely changed in the salted and dried yellow corvenia. During the processing and storage of Gulbi, DMA and TMA contents were less produced in brine salted and dried yellow corvenia using crude salt than in sample prepared using pure salt, while the former were more effective than the latter in inhibiting the production of nitrate and nitrite. Therefore, it was revealed that reduction of NA precursors such as DMA, TMA, nitrate and nitrite were more effective in preparing with the brine salting method than with the dry salting method.

• 정보처리학회논문지B
• /
• 제13B권4호
• /
• pp.429-438
• /
• 2006

일반적으로 신경망은 비선형성 문제를 해결하기 위해서 소프트웨어로 많이 구현되었지만, 영상처리 및 패턴인식과 같은 실시간 처리가 요구되는 응용에서는 빠른 처리가 가능한 하드웨어로 구현되고 있다. 다양한 종류의 신경망 중에서 다층 신경망(MLP: multi-layer perceptron)의 하드웨어 설계는 빠른 처리속도와 적은 면적 그리고 구현의 용이성으로 고정소수점 연산을 많이 사용하였다. 하지만 고정소수점 연산을 사용하는 하드웨어 설계는 높은 정확도의 부동소수점 연산을 많이 사용하는 소프트웨어 MLP를 쉽게 적용할 수 없는 문제점을 가진다. 본 논문에서는 높은 정확도와 높은 유연성을 가지는 부동소수점 연산을 사용하면서도 은닉층 뉴런수를 주기(cycle)로 빠르게 수행하는 MLP의 완전 파이프라이닝(fully-pipelining) 설계방법을 제안한다. MLP는 주어진 문제에 의해서 자연스럽게 입력층과 출력층의 구조가 결정되지만, 은닉층 구조는 사용자에 의해서 결정된다. 그러므로 제안된 설계방법은 많은 반복수행이 요구되는 영상처리 및 패턴인식 등의 분야에서 은닉층 뉴런수를 최적화 하여 쉽게 성능 향상을 이룰 수 있다.

• 디지털융복합연구
• /
• 제12권11호
• /
• pp.379-386
• /
• 2014

본 논문에서는 Grabcut 알고리즘의 수행 속도를 효율적으로 개선시키기 위하여 GPU(Graphics Processing Unit)에서 데이터를 처리하는 방법을 제안한다. Grabcut 알고리즘은 뛰어난 성능의 객체 추출 알고리즘으로 기존의 Grabcut 알고리즘은 전경 영역과 배경 영역을 분할한 후 배경 K-클러스터와 전경 K-클러스터로 할당한다. 그리고 할당 된 결과를 점진적으로 개선될 때까지의 과정을 반복한다. 하지만 Grabcut 알고리즘은 반복된 클러스터링 작업으로 인하여 수행 시간이 오래 걸리는 단점이 존재한다. 따라서 GPGPU(General-Purpose computing on Graphics Processing Unit)를 이용해 반복되는 작업을 병렬적으로 처리하여 Grabcut 알고리즘의 수행 속도를 효율적으로 개선시키는 방법을 제안한다. 제안하는 방법으로 Grabcut 알고리즘의 수행시간을 평균 약 90.668% 감소시켰다.

• 대한임베디드공학회논문지
• /
• 제9권6호
• /
• pp.345-351
• /
• 2014

This paper presents pedestrian recognition to improve performance for vehicle safety system or surveillance system. Pedestrian detection method using HOG (Histograms of Oriented Gradients) has showed 90% recognition rate. But if someone takes a picture in the rain, the image may be distorted by rain streaks and recognition rate goes down by 62%. To solve this problem, we applied image decomposition method using MCA (Morphological Component Analysis). In this case, rain removal method improves recognition rate from 62% to 70%. However, it is difficult to apply conventional image decomposition method using MCA on vehicle safety system or surveillance system as conventional method is too slow for real-time system. To alleviate this issue, we propose a rain removal method by using low-pass filter and DCT (Discrete Cosine Transform). The DCT helps separate the image into rain components. The image is removed rain components by Butterworth filtering. Experimental results show that our method achieved 90% of recognition rate. In addition, the proposed method had accelerated processing time to 17.8ms which is acceptable for real-time system.